TRIUMF: Canada's national laboratory for particle and nuclear physics

Annual report scientific activities, 1986 TRIUMF Aug 31, 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


51833-TRI-Annual Report-1986.pdf [ 117.17MB ]
JSON: 51833-1.0107789.json
JSON-LD: 51833-1.0107789-ld.json
RDF/XML (Pretty): 51833-1.0107789-rdf.xml
RDF/JSON: 51833-1.0107789-rdf.json
Turtle: 51833-1.0107789-turtle.txt
N-Triples: 51833-1.0107789-rdf-ntriples.txt
Original Record: 51833-1.0107789-source.json
Full Text

Full Text

TRIUMFANNUAL REPORT SCIENTIFIC ACTIVITIES 1986MESON F A C I L I T Y  OF:U N I V E R S I T Y  OF ALBERTA S IMON FRASER U N I V E R S I T Y  U N I V E R S I T Y  OF V I C T O R I A  U N I V E R S I T Y  OF B R I T I S H  COLUMBIAOPERATED UNDER A CONTRI BUT ION FROM THE NATIONAL  RESEARCH COUNCIL  OF CANADA AUGUST 1987TRIUMFANNUAL REPORT SCIENTIFIC ACTIVITIES 1986TRIUMF4004 WESBROOK MALL VANCOUVER, B.C. CANADA V6T 2A3E X IS T IN GPROPOSED REMOTEHANDLIFACILirPROTON HALL EXTENSIONSERVICEBRIDGESERVICE'ANNEXEXTENSION H ION SH POLAR ION SCHEMISTRYANNEXMESON HALLM 9 ( H / j j )  M 2 0 ( jj )42 MeV ISOTOPE PRODUCTION CYCLOTRONMESON HALL EXTENSIONINTERIMRADIOISOTOPELABORATORYr ELINGTY>)L 2 C  I L 1 A  ( P )iOURCERIZEDiOURCEBAT HOBIOMEDICALLABORATORYTHERMALNEUTRONFACILITYMESON HALLSERVICEANNEXFOREWORDIt is a pleasure to introduce the 1986 Annual Report of scientific activities of TRIUMF, a project which remains a strong contributor to the world effort in subatomic physics and a leading window on international science for Canada. It was an interesting year in which a great deal of important science emerged, as chronicled in this report. Also, it was the first year in more than a decade in which the project did not experience real growth in its operating funds. TRIUMF contributed to a government-wide program of fiscal restraint. Measures taken for this purpose occurred abruptly at the beginning of the year and TRIUMF responded quickly emerging, we believe, with a stronger relationship with its funding agency in Ottawa, the National Research Council.During 1986 TRIUMF began the vigorous pursuit on its major update project - the KAON Factory. It requires courage to be optimistic in times of fiscal restraint. It is a credit to TRIUMF that this major vision for the future occurs at the prime of its life when it has before it a very healthy program of ongoing activities covering a wide spectrum of science.The essential feature of TRIUMF is that it undertakes world-class physics with a mixture of people and ideas from around the globe. TRIUMF began as a major regional centre of excellence in Western Canada, but has now emerged as Canada's national facility for sub­atomic physics. Graduate students and scientists from across the land are meeting its scientific challenges and developing its new research programs. Its management framework as a joint venture of a consortium of universities together managing a major national enterprise remains a significant model for Canada. We can be optimistic for TRIUMF as long as its program remains as strong as the scientific output chronicled in this report.P.A. LarkinChairman, Board of ManagementvTRIUMF was established in 1968 as a laboratory operated and to be used jointly by the University of Alberta, Simon Fraser University, the University of Victoria and the Uni­versity of British Columbia. The facility is also open to other Canadian as well as foreign users.The experimental programme is based on a cyclotron capable of producing three simultaneous beams of protons, two of which are individually variable in energy, from 180-520 MeV, and the third fixed at 70 MeV. The potential for high beam currents — 100 pA at 500 MeV to 300 pA at 400 MeV — quali­fied this machine as a 'meson factory'.Fields of research include basic science, such as medium- energy nuclear physics and chemistry, as well as applied research, such as isotope research and production and nuc­lear fuel research. There is also a biomedical research facility which uses mesons in cancer research and treatment.The ground for the main facility, located on the UBC campus, was broken in 1970. Assembly of the cyclotron started in 1971. The machine produced its first full-energy beam in 1974 and its full current in 1977.The laboratory employs approximately 363 staff at the main site in Vancouver and 18 based at the four universities. The number of university scientists, graduate students and support staff associated with the present scientific pro­gramme is about 460.viCONTENTSINTRODUCTION j_SCIENCE DIVISION 3Introduction 3Particle Physics 5Measurement of the n-p spin correlation parameter Ann 5Measurement of the slope of the it0 electromagnetic form factor 5A study of the decay tt ev 7Radiative muon capture with the TPC 8Test of charge symmetry by the comparison of da(Tt+d -*• pp) withda(ir-d * nn) 9Measurement of parity violation in p-p scattering 1 1Energetic neutron spectrum from \T capture in deuterons 11The reaction ppir0 near threshold 1 3Muonium-antimuonium conversion 14Ratio of spin transfer coefficients 1 7Measurement of K+ ->• ir+v'v 18SLD group 20Nuclear Physics and Chemistry 22Proton-proton bremsstrahlung 22Study of simple features of the A(p,ir~)A+l reaction in the3,3 resonance region 22Validity of DWBA analysis for intermediate energy proton scatteringto low-lying states of 208Pb 24Initial studies of the (n,p) reaction on light nuclei 26The 2 0 8Pb(n,p) reaction at 200 MeV 26Gamow-Teller strength in 21tMg from spin-flip cross sections ininelastic proton scattering 26Inelastic scattering of 30 and 50 MeV tt+ projectiles from thestate of 12C 29Investigations of the pion absorption reaction 6Li ,1 2C( tt^X 1X2)A 29Elastic and inelastic proton scattering from 208Pb at medium energies 30Measurement of differential cross sections at T^ = 90 MeV 31Giant resonance study with 75 MeV tt~ 32Polarization analysing power differences for inelastic protonscattering from 12C at 400 MeV 32Study of the (tt+, tt+tt-) reaction on 160, 28Si and 1+0Caat 1^ = 240 and 280 MeV 3 44" stretched states in 160 36Energy dependence of the N-nucleus interaction from 1+ excitationsin 2 8Si(p,p) 36Measurement of tensor observables in ttJ elastic scattering 3 9Excitation of stretched particle-hole states in chargeexchange reactions 41Study of the energy dependence and the A dependence of the doublecharge exchange reaction at low energy 41Study of nuclear structure and density dependence of the effectiveinteraction for N=50 isotones 43Measurement of Bq^ for and 28Mg using the (n,p) reaction 43The spin response for 5‘tFe(p,p') at 290 MeV 44A search for the tetraneutron 46Measurement of da/dfl and A^q to exclusive states of 180(p, tt+) 170between 250 and 500 MeV 4 7Low energy pion scattering and pionic atom anomaly 48Non-analog DCX, the 160(tt+, it- ) 16Ne reaction 48viiElectroweak interactionsAstrophysicsMuon spin rotationIntroduction Beam production CyclotronCyclotron development4950 5049Nuclear Physics and Chemistry (cont'd)Study of the 90Zr(n,p) reaction at 200 MeV Measurement of the charge symmetry parameter Aw in the ird elastic scattering reaction Energy dependence of the (p,n) cross section for C and N Measurement of spin observables using the (p ,p 'y) reaction A test of the Gamow-Teller sum rule using charge exchangereactions on 59Fe 51Pion absorption reactions 6Li , 12C( tt+ ,X1,X2)A 54Research in Chemistry and Solid-State Physics 55Temperature dependence of reaction rate constants for muonaddition reactions in liquid phases 55Diluted magnetic semiconductors 55Quantum diffusion of muons 57Gas phase muonium addition reaction kinetics 58Kinetic isotope effects in the Mu+H2 and Mu+D2 reactions 58Muon molecular ions and ion—molecule reactions 59Spin dynamics in crystalline and amorphous DyAg 60Thermally activated muonium formation in BaF2 and A£203 61yLCR spectroscopy of free radicals 62Muonium in water ^4High pressure muon spin resonance in liquidsResolved nuclear hyperfine structure of anomalous muonium insemiconductors ^Muonium in micellesMuon polarization in Xe up to 5 atmMuon spin rotation studies of supported metal catalysts Hyperfine structure of muon and muonium defect centres in magnetically ordered fluorides6667686970707072Theoretical Program IntroductionScattering and reactions on nuclei Nuclear structureScattering and reactions on nucleons 72Baryon structure 77Hadron structure/quark model 78Lattice gauge calculations 79Grand unified models 7^808081APPLIED PROGRAMS DIVISION 8282Introduction O OBiomedical program42 MeV cyclotron 82Radioisotope processing (AECL) 8^500 MeV radioisotope productionPositron emission tomography (PET) 88TRIM program 88CYCLOTRON DIVISION 87878990 90viiiCYCLOTRON DIVISION (cont'd)Cyclotron (cont'd)rf operation 95Vacuum upgrade 96Diagnostics upgrade 96Engineering physics 96Beam quality 96Projects 98Alternative extraction system 98Resonator improvement program 99New projects 99Ion sources and injection system 101Primary beam lines 103Control system 105Operational services 107EXPERIMENTAL FACILITIES DIVISION 109Introduction 109Experimental support 110Data acquisition systems 110Nucleonics and IAC 110Detectors facility 111MWPC facility 112Meson hall 112M9 channel upgrade 112Mil channel 113M13 channel 114M15 channel 114QQD spectrometer 114Proton hall 115Beam line 4B 115MRS 116Waterfall target for high resolution proton inelastic 160scattering experiments at the MRS 116TISOL facility 117Dual arm spectrometer system/second arm spectrometer (DASS/SASP) 118Targets 120Experimental facilities engineering 121ACCELERATOR RESEARCH DIVISION 124Introduction 124Beam development 125Cyclotron 125ISIS 125Central region 125Effect of rf booster cavities 125Changes in cyclotron characteristics 126Alternative extraction program 126Primary beam lines 128Complete longitudinal polarization in the proton hall 128A proton radiotherapy beam line at TRIUMF 128Secondary channels 129Superconducting solenoid muon channel 129Mil pion channel 1301AT2 upgrade 130Stopped muon channel at LAMPF 130MUS0L 130Beam line diagnostics 130ixACCELERATOR RESEARCH DIVISION (cont'd)Computing services 132KAON factory studies 133TECHNOLOGY AND ADMINISTRATION DIVISION 143Introduction 143Site services 143Safety program 143Building program 145Mechanical engineering 146Planning 146Controls, electronics and computing 147Electronics 147Data Analysis Centre 148Administration 151Business Office 151CONFERENCES, WORKSHOPS AND MEETINGS 152ORGANIZATION 154APPENDICESA. Publications 157B. Users group 169C. Experiment proposals 172xINTRODUCTIONIn the world network of large international accelerator facilities TRIUMF is extraordin­ary in its variety of beams and range of science, in its consequent ability to respond to the rapidly changing science needs of the field of subatomic physics and in its produc­tion of spin-offs. The portrait of TRIUMF's science in this annual report for 1986 is intended to convey that image of a project now in the fullest vigour of its scientific life. It is essential to view the portrait as a whole, the strength of the laboratory resides even more in the combination of its many activities and programs than in the individual merit of the various experiments.The driving force for subatomic physics, worldwide, remains the new picture of basic building blocks (quarks and leptons) and of a unified description of the fundamental forces. Many experiments at TRIUMF have con­tributed importantly, in the past few years, to achieve this new understanding of what lies at the heart of matter. Both the present TRIUMF program and the major new TRIUMF ex­pansion plans provide Canada with the oppor­tunity to be a major player in the response to the many important questions which have emerged from this new understanding.This annual report pertains primarily to the present TRIUMF program and only briefly to the developments for the TRIUMF Upgrade - the KAON factory. In 1986 the KAON factory grew as a major preoccupation of TRIUMF. It is a vision for the future which must be kept in mind in this portrait of the present.The question which should be in the mind of the reader of this portrait of TRIUMF's activities in 1986 is: how well is TRIUMFusing its special advantages and tools in addressing the important science issues? The work of TRIUMF extends considerably beyond subatomic physics - as we note below and as can be learned from the contents of this report - but taken as a whole, TRIUMF's con­tributions to subatomic physics are central in answering this question.Any assessment of TRIUMF's contribution to subatomic physics must note the variety of TRIUMF's beams and research facilities, in contrast to the very large collider facili­ties in a number of other countries. The colliders work at the energy frontier with much greater energy than is contained in theprimary proton beam of TRIUMF. Each collider facility, in general, has a single experiment in which spectacular collisions produce directly new sought-after basic building blocks. For example, in 1983 _the proton- antiproton collision of the SPPS facility at CERN in Switzerland first produced the W"*", W- and Z° particles which are partners to the quantum of light (the photon) in the new theory unifying electromagnetism with the weak nuclear force responsible for radio­active decay. This experiment was one of the largest and perhaps the most important - in our century of science and immediately won the Nobel prize for its leader, Carlo Rubbia. TRIUMF scientists participated in that ex­periment and Alan Astbury, now of TRIUMF and the University of Victoria, was the co-leader of that discovery. For direct drama the colliders cannot be excelled.TRIUMF approaches the same science, and more, with experiments using lower energy beams but of much greater intensity. For example, in the same year that the W1- was discovered at CERN it was at TRIUMF that its most important property was discovered - namely its left- handedness. Loosely speaking it behaves like a left-handed corkscrew rather than a right- handed one. The W-*" is associated with the decay of the positively charged muon and the right-handedness governs the polarization of the electron emitted in muon decay. It was TRIUMF's pure and intense beams of low energy muons which made possible the elucidation of the handedness. This example illustrates how the accelerators, such as TRIUMF, providing extraordinary intensity complement the science of the high-energy colliders.TRIUMF provides pions, muons, protons and neutrons as tools for subatomic physics. As the questions change the interest shifts to experiments with different tools. For variety of program and opportunity, the meson factory cannot be excelled.In TRIUMF's early years, a decade ago, experiments involving the break-up of the muon and the pion were on centre stage. In TRIUMF's meson hall the focus was on beam lines and other experimental facilities for this purpose. In 1986 some major experiments of this kind were completed, for example, the six-year search for neutrinoless muon conver­sion with TRIUMF's superb time projection chamber. A major contribution has been madein TRIUMF's meson hall toward the understand­ing of muon and pion decay. After the first frontal attack on those experiments there is likely to be somewhat of a pause now for several years until advances in detector technology demand a new look. In the meantime the interest in the meson hall will shift to other experiments and a larger fraction of TRIUMF's particle physics effort will involve the preparation of experiments to be mounted at other laboratories abroad. The very in­tense effort on the rare-kaon experiment at Brookhaven and the SLD experiment at Stanford (both described in this report) are examples of this shift in interest.For the past five years the development of experimental facilities at TRIUMF has shifted from the previous focus in muons to new opportunities with proton and neutron beams in the proton hall. The interest here is in the structure of the atomic nucleus and especially on the new questions raised by the new picture of quarks and fundamental forces. TRIUMF, with its high-quality variable-energy and polarized proton beams, had unique oppor­tunities and it was the right time for this science to emerge onto centre stage. What has emerged, as described in this report, is a major new operating nuclear physics facility unique on the world scene. With its spectrometer, polarimeters, charge-exchange and spin capabilities this combination of facilities is producing exciting science and promises to do so for a decade.In the meson hall the subatomic physics interest is now likely to turn to using the existing secondary beam lines - formerly so important for muon break-up experiments - for pion-nucleus experiments. New facilities, combinations of spectrometers and multitrack detectors, will be needed for this science. The meson-nucleus interaction was a very important original science motivation of the world's three meson factories two decades ago. In the meantime muons and protons intruded, although both at TRIUMF and its sister meson factories there have been impor­tant preliminary skirmishes on the pion- nucleus interaction, especially on elastic scattering. The new picture of subatomic building blocks and forces has raised new questions for which pion-nucleus studies now appear to be ready to assume centre stage. Moreover, the technology of detectors and spectrometers has evolved so that one can nowthink of building devices whose acceptance and resolution largely compensate for the intrinsic low flux of any secondary beam line. It appears that the time is now finally ripe for this program.With subatomic physics in the meson hall in transition one should note, in this report, the very major condensed matter and pion- therapy program in the same hall. It is very satisfying to have now the enhanced survival data for patients with deep-seated tumours treated with pions, as given in this report. The ySR studies (muon spin precession) cover an extraordinary range of studies of chemis­try and physics in gases, solids and liquids. Of special importance in 1986 were the studies in gas kinetics and the level- crossing experiments with free radicals. This work now involves two of TRIUMF's secondary beam lines in the meson hall and will soon have a third.TRIUMF's cyclotron performed well in 1986. It is a large complex tool subject constantly still to development and improvement. It was a very significant event for TRIUMF in 1986 to understand the nature of the rf leakage in the TRIUMF resonators and to then be able to reduce it by an order of magnitude. This breakthrough improves the cyclotron vacuum, greatly reduces losses of beam in the cyclo­tron and will lead directly to the achieve­ment of significantly higher beam currents for meson production.The TRIUMF applied program is growing rapidly as one would expect for such a diverse project in its first full scientific vigour. Initial spin-offs came from the challenge of building the world's largest and most complex cyclotron. The present spin-offs are in ideas, people and patents for medicine, electronics,etc. The sale of TRIUMF-produced isotopes appears to be doubling annually and reached several million dollars in 1986. The use of TRIUMF's PET brain scanner produced many important results, especially the study of patients exposed to the neurotoxin MPTP whose effects simulate some of those associ­ated with Parkinson's disease. These results received significant attention international­ly in the public TV media. It is now the right time for TRIUMF to focus more effort on the commercialization of ideas emerging from its varied research program.2SCIENCE DIVISIONINTRODUCTIONTRIUMF had another very productive year in 1986, with significant and important experi­ments carried out in particle physics, nuc­lear physics, chemistry and solid state physics.The cyclotron continued to operate reliably, although the total number of milliampere- hours delivered to experiments was slightly less than last year, being just over 300. The reduction was mainly because we ran pol­arized beam for about one week longer than last year (and unpolarized beam for one week less) to accommodate the tremendous pressure for polarized beam, where requests are typi­cally a factor of three times availability. Experiments with polarized beam are run with beam current more than two orders of magni­tude lower, and so polarized running contrib­utes a negligible amount to the integrated beam delivered.There were some changes in the demand for the different beam lines at TRIUMF, reflecting new facilities and changing scientific prior­ities. In the proton hall pressure for pol­arized beam continued to be extraordinarily high, as noted above, while there was a great increase in demand for unpolarized beam. This has occurred because of the commission­ing in late 1985 of the new CHARGEX facility, which enables experiments on (p,n) and (n,p) reactions to be carried out for the first time at intermediate energies. This facility is truly unique in the world and will remain so for the next few years. It has attracted great interest in the international communi­ty, as was evident when the Nuclear Physics Divisional Meeting of the American Physical Society was held in Vancouver in September. The meeting was preceded by a workshop on (p,n)(n,p) reactions attended by the leading luminaries in the field. By the end of the year data-taking had been completed on about half a dozen of the first experiments and the first publications are beginning to appear.In the meson hall the major change has been a shift in emphasis from M13, the low-energy pion line, to Mil, the highest energy pion line at TRIUMF. Over the last two years there has been a great improvement in the under­standing and performance of Mil. In addition, the building and commissioning of the polar­ized deuteron target means that a full pro­gram of pion-deuteron vector and tensor polarization measurements can now be carried out, and this program is in full swing. The declining popularity of M13 reflects the fact that many of the experiments which can be carried out with the resolution of the QQD (800 keV-1 MeV) are now complete, and a new facility is needed to keep TRIUMF in the forefront of pion-nuclear physics. The physi­cists who work in this area are actively pursuing such a development. M15 and M20 continue to be in great demand for pSR ex­periments. TRIUMF has pioneered the develop­ment of a new technique, level crossing resonance, which promises to be of great importance in the application of ySR to both solid-state physics and chemistry.Turning to some of the highlights of the ex­perimental program, the most important mea­surement in particle physics in 1986 was Expt. 248, a study of the decay tt + ev. The branching ratio for this decay provides the most direct and stringent test of universal­ity in weak interactions, and is sensitive to the presence of pseudoscalar interactions. The experiment also sets tight limits on the possible existence of massive neutrinos. Data-taking on M13 was completed during the first half of 1986, with about 3xl05 n+ev decays and 2x108 ir-vp+e decay events recorded on tape. The experiment should decrease the errors on the previous best measurement (also from TRIUMF) by a factor of 3.In January the last run of the y-e conversion experiment (104) took place on the time pro­jection chamber, completing several years of data-taking, and setting an upper limit of 5xl0-12 for this important decay. The TPC was then used for measurements of radiative muon capture on calcium and carbon, which are particularly sensitive to the induced pseudo­scalar form factor of the hadronic current (Expt. 249).TRIUMF continued its contributions to the study of the nucleon-nucleon interaction with measurements in January of the ratio of the3Wolfenstein parameters Rt/Dt, and the frozen spin target was refurbished in preparation for a measurement of the spin correlation parameter Ayy in n-p scatter­ing, which is planned for early 1987.The MRS continued to be the centrepiece of the nuclear physics program in the proton hall. This versatile instrument, now equipped with a focal plane polarimeter and also equipped for its new role as the recoil spec­trometer in the CHARGEX facility, was in almost continuous use throughout the year. Whilst it is difficult to pick any one exper­iment above the others, we can take pride in the extensive series of measurements on ^Fe, which included cross sections, analysing powers, spin response and (p,n) and (n,p) experiments, and set a new standard for com­prehensive measurements on a single nucleus.In the meson hall the important work in understanding pion double charge exchange continues, and the first measurements of the pion-induced pion production were made uti­lizing a second arm plastic scintillator pion detector in coincidence with the QQD.Amongst the chemistry experiments which form part of the ySR program, two stand out. Experiment 339 studied kinetic isotope effects in Mu+H2 and Mu+D 2 reactions, for which accurate ab initio calculations can be made, and provided accurate data to test reaction rate theory. Experiments 358 and 398 applied the level crossing resonance technique to the study of free radicals. The same LCR spectroscopy was applied in Expt. 367 to study the nuclear hyperfine structure of anomalous muonium in GaAs and GaP, and has led to a great increase in understanding of this important problem.The year finished on a happy (if busy) note with the December meeting of the Experiments Evaluation Committee receiving 39 new pro­posals, an increase of one-third over any previous meeting. We take this as evidence that TRIUMF is a world-class facility in whose continuing vitality the country can take pride.The contributions on individual experiments in this Report are out­lines intended to demonstrate the extent of scientific activity at TRIUMF during the past year. The outlines are not publications and often contain preliminary results not intended, or not yet ready, for publication. Material from these reports should not be repro­duced or quoted without permission of the authors.4PARTICLE PHYSICSExperiment 182Measurement o f the n-p spin correlationparam eter Ann(W.T.H. van Oers, W.D. Ramsay,Manitoba)This experiment is to measure the spin corre­lation parameter Ann in n-p elastic scatter­ing to an absolute accuracy of ±0.03 at 220, 325, 425 and 495 MeV. The measurement con­sists of scattering a polarized neutron beam from a polarized proton target and measuring the difference in yield when the spins are changed from parallel to antiparallel. An accurate measure of Ann at the energies proposed will make substantial improvements to the nucleon-nucleon data base. It will provide an important constraint on isospin zero nucleon-nucleon phase-shift solutions, will resolve some discrepancies in the data base and will help to understand differences between predictions of the Paris and Bonn potentials.To be of maximum use the experiment requires good absolute accuracy. High statistical pre­cision is not enough. The target polariza­tion, for example, must be known to an abso­lute accuracy of 2%. Since the target pola­rization is measured by an NMR system with an absolute calibration no better than 4%, an independent absolute calibration will be needed. The group plans to do this by scat­tering protons from the target at an energy for which the pp analysing power is accurate­ly known. The proton beam will be produced by replacing the liquid deuterium in the neutron production target with liquid hydro­gen and using the protons scattered through the 9° collimator port through which the neutron beam would normally emerge. During target calibration with this secondary proton beam the clearing magnet (see Fig. 1) will be turned off and the primary proton beam dumped in a beam stop inserted in the 0° collimator port.During the past year tests were made to de­termine the quality of the secondary proton beam. Its energy spread was determined by measuring the time of flight between two small fast plastic scintillators (Pilot-U).A 1 cm cube was placed just after the 9° collimator. A second scintillator, a 2.54 cm diameter by 1 cm thick disc, was located farther downstream. With 1 m detector sepa­ration the proton peak was 150 ps FWHM. At 9.75 m detector separation the proton peakbroadened to 390 ps FWHM, indicating a beam energy spread of 14 MeV FWHM. When the air in the 9° collimator was replaced with helium, the energy spread was reduced to 12 MeV. Since the variation in pp analysing power with energy is only 5xl0-4 per MeV at 498 MeV, the error in our calibration intro­duced by beam energy spread will not be significant.During polarization calibration runs the pro­ton hit position on target will be determined by a pair of fast drift chambers of the type used on the front end of the TRIUMF medium resolution spectrometer. The location of these wire chambers is shown in Fig. 1. These chambers have been tested in the secondary proton beam and it should be possible to de­termine the hit position on target to ±1 mm.Much work has also been done on optimizing the performance of the frozen spin target (FST) and its NMR polarization measuring sys­tem. A temperature of 40 mK is now obtainable on a routine basis. This gives a polariza­tion decay time constant in excess of 500 h. Slow polarization decay is very important as the absolute calibrations will be time con­suming and should be done as infrequently as possible. The NMR system has been improved and is now repeatable to within about 2% over a one-week period. This is also very impor­tant as the NMR must be trusted between abso­lute calibrations.A test run and the first production run have been scheduled for early 1987.Experiment 217Measurement o f the slope o f the ir° electromagnetic form factor (J.-M. Poutissou, TRIUMF; A M  Stetz,Oregon State)We have measured the slope parameter of the it0 electromagnetic form factor 'a' by mea­suring the partial branching ratio of the Dalitz decay ir° ->- ye+e~ into high invariant mass electron-positron pairs.The experiment carried out in 1984 on the M13 channel was described previously (1985 annual report, p. 10). So far we have analysed the data taken at 60° and 130° for the pairs opening angle. The 60° data are not sensitive to the form factor and are used to confirm the overall normalization of our Monte Carlo5Fig. 1. Layout of equipment for the Ann measurement.60.0 0.2 0.4 0.6 0.8INVARIANT MASS X visFig. 2. The invariant mass distribution Xv^g at 130°. The data are represented by thehistogram; the Monte Carlo result is thesolid curve.simulation. Figure 2 gives a comparison of the invariant mass spectrum measured (histo­gram) with the prediction of the Monte Carlo (line). No parameters are adjustable in this f it.We have estimated the corrections due to radiative processes up to second order and confirmed that we are much less sensitive to them than previous experiment (-0.01±0.01). Our systematic error estimates are still con­servative at this stage, but we can say that we are not compatible with the previous large value for 'a'. We observed a value for 'a' which is not in disagreement with the vector dominance estimates. Our result is:a = -0.01 ± 0.035+ 0.07 - 0.05Experiment 248 A study o f the decay ir— ev (T. Numao, TRIUMF)The measurement of the branching ratio R = (ir+ev)/(ir+pv) provides a stringent test of universality in weak interactions. Because the decay n + ev is helicity suppressed by a factor of lO4, R is also sensitive to the presence of pseudoscalar interactions intro­duced by charged Higgs particles and lepto- quarks. The standard model assuming univer­sality now predicts R = 1.234xl0-1+ with a negligible uncertainty owing to the recent experimental progress on radiative pion decays [Bryman et al., Phys. Rev. D 33^ 1211(1986); Egli et al., Phys. Lett. 175B, 97(1986)]. The most accurate branching ratio measurement was R = (1.218+0.014)xl0_It, which is in agreement with the theoretical predic­tion [Bryman, op. cit.]. The goal of the present experiment is to improve the accuracy of the branching ratio by a factor of 3.In addition to the test of universality the same experiment sets tight limits on the pos­sible existence of massive neutrinos [Azuelos et al., Phys. Rev. Lett. 56_, 2241 (1986)]and other new particles such as majorons. The presence of massive neutrinos could not only affect the branching ratio but also introduce additional peaks in the positron energy spec­trum through neutrino mixing. Since helicity suppression is not applicable to heavy neu­trinos, the search for additional peaks in the -rr+ev spectrum can be very effective.The experiment was carried out on the M13 channel using an 85 MeV/c rr+ beam. Positive pions, stopped in a scintillator target, de­cay either to e+ (Eg+ = 70 MeV) throughthe decay ir+ev or to p+ through the decay ir+pv followed by the decay p-evv (ir-p-e chain, Eg+ = 0-53 MeV). Positrons from both decays were detected with a 51 cm diam x 46 cm Nal (TL) detector TINA, which was used in con­junction with a telescope consisting of 2 wire chambers and 4 thin plastic counters, as shown in Fig. 3. About 3xl05 ir+ev decay and 2xl08 ir-p-e chain-decay events were recorded on tape.For the search of massive neutrinos in the decay Tr+ev, large background from the ir-p-e decay chain was suppressed by accepting onlyB5Beam B1 B 2 -4/ XT 1— 3CW1 W2-T4TINAFig. 3. Schematic of the experimental setup. B, T and W indicate beam counters, positron telescope counters and wire chambers. For clarity, several beam-defining counters are not shown in the figure.7early decay events and by inspecting the total energy deposited in the target counter. The present data set includes 10 times more TT+ev events and the background suppression factor has been doubled over the one in the recent publication from our test run [Azuelos et al., op. cit.], which will result in significant improvement for the mixing matrix i U e i I 2 in the neutrino mass region 60-130 MeV/c2 .The branching ratio is calculated from the numbers of events in the 70 MeV peak and in the iT-y-e decay distribution with some cor­rections. The largest uncertainty in the previous experiment was due to a lack of precise knowledge of the low energy tail of the Nal line shape. In the present work this is empirically determined from a background- suppressed TT-*ev spectrum. The data analysis is in progress.Experiment 249Radiative muon capture with the TPC(G. Azuelos, TRIUMF)Radiative muon capture (RMC) is a weak semi- leptonic process which is particularly sensi­tive to the induced pseudoscalar form factor gp of the hadronic current. A programme is under way at TRIUMF to measure RMC on light nuclei to investigate its possible renormali­zation in the nucleus. Measurements have been made of RMC on calcium and carbon.The TRIUMF TPC is used as a large solid angle (2.5 sr), medium resolution (10%) magnetic pair spectrometer to detect y-rays from RMC. The photon acceptance over the energy range of interest is measured using y-rays from tt° decay from n-p ir°n, obtained by asubtraction suitably normalized of spectra from it-  stopping in CH2 and C. The measured spectrum (using a 1.0 mm Pb photon converter) is compared with a Monte Carlo calculation in Fig. 4. The decrease in acceptance at lower energy is due to our trigger requirement that both the e+ and e” are sufficiently energetic to reach the outer trigger counters. A sig­nificant dependence of the measured accept­ance on the particle flux in the chamber was found, likely due to space charge and posi­tive ion effects. At low rates the average acceptance in the region of the tt° spectrum, including track reconstruction and software cuts, was 0.26% with a 1.0 mm thick converterRMC measurements in the past have been typi­cally plagued by backgrounds due to neutronsGamma Ray Energy (MeV)Fig. 4. y-ray spectrum from ir° decay, used to determine the detector acceptance (A) com­pared with Monte Carlo calculation (histogram).and radiative pion capture, neither of which is a difficulty with the present experiment. The TPC is by its nature insensitive to neu­trons, and pions are rejected by an rf separator and by the rejection of any photon event in prompt coincidence with a signal in the beam counters. The overall pion rejec­tion efficiency was measured to be better than 4xl0-7, the remaining pion-induced events contributing <1% to the observed RMC rate.Data were taken on Ca and C, each with both a0.6 mm and a 1.0 mm lead converter. Figure 5 shows the 1.0 mm converter data for Ca and C, respectively. Only the energy range >57 MeV is usable due to the background from radia­tive muon decay and bremsstrahlung of Michel electrons in the target. The total number of good y-rays of energy >57 MeV from the Ca target is about 2000, divided equally between runs with the two thicknesses of converter, and about 1100 from the C target, mostly taken with the thicker converter. Comparison of the results of the measurements with dif­ferent converters will provide a valuable check on the systematic errors. A measure­ment of RMC on oxygen using a D,,0 target is also in progress.Progress has been made on the design of a drift chamber (based on the chamber for the Brookhaven Expt. 787) for the measurement of RMC on hydrogen, using Monte Carlo routines based on the GEANT program. Various possi­bilities for measurement of the z-coordinateGamma Ray Energy (MeV) Gamma Ray Energy (MeV)Fig. 5. RMC spectrum from (a) Ca, (b) C, with 1.0 mm Pb converter.of the tracks have been investigated, includ- to reduce the background, so that the numbering charge division, stereo wires, inner and/ of singles events may be determined more ac-or outer wire chambers and helical inner curately for the in situ, precision calibra-scintillators. A single-cell prototype tion of the neutron counter efficiency. Inchamber has been constructed and testing is the current analysis three cuts were made onunder way. The measurement on hydrogen with the neutron TOF spectra: 1) beam TOF, 2) beamthe drift chamber has been submitted to the pulse height, and 3) neutron pulse height.EEC as a separate proposal (Expt. 452). The cuts reduce the background significantlywhile very few peak events are lost. The TOF spectra are shown at each stage of the analy- Experiment 270 sis in Fig. 7(a-d). Note that the beam tim-Test o f charge sym m etry by the comparison o f ing and beam pulse height cuts largely reducedcr(ird — pp) with d o ( ifd — nn) the slow neutron peak, while the neutron(B.M.K. Nefkens, UCLA) pulse height cut reduces the general back­ground much more uniformly. In Fig. 7(d) the Four angles in the range 20° < 9c.m. * ^0° target-empty background run has been sub-were measured simultaneously at four beam tracted. Unlike the protons there is still aenergies, 142, 180, 217 and 254 MeV. These residual background ranging from 10-20%.energies were chosen to cover the A intermed­iate state. High statistics runs were taken at the first three energies; the running at 4,0the highest energy was exploratory. The set­up was described in detail in last year's annual report.3.0The proton data have all been replayed. The absolute cross sections at 142 MeV are shown &  in Fig. 6 and compared with results obtained by Richard-Serre et al. [Nucl. Phys. B20, 413 <0 2.0 (1970)] and Boswell et al. [Phys. Rev. C 25,2540 (1982)]. Our experiment provides a nice test of detailed balance (at the 3% level) by a comparison of our cross sections with those ,0 of the inverse reaction pp > w+d obtained by Hoftiezer et al. [Nucl. Phys. A402, 429(1983)].01 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0The ir~d ■+ nn data are analysed in much the Cos’Ocmsame way as ir+d + pp. A sequence of cuts isapplied to the time-of-flight (TOF) spectra Fig. 6. Test of detailed balance.M2 MeV o taTT d -> p p Ida/a  n  hn,b/<r t ° r  Ib 10 xai ■ 1BK.rX  EXP 270 •  C  Rlchard-S«rr« O Boiwetl O HoftleierJ_______I_______I_______L_90Fig. 7. Left side neutron TOF singles spectrum (a) no cuts, (b) beam timing and beam pulse height cuts, (c) neutron pulse height cut and previous cuts, and (d) all cuts and target-empty yield subtracted.Preliminary data analysis gives the followingneutron counter efficiencies at T,counter 1 counter 2 counter 3 counter 4iR0.20±0.01 0.23+0.01 0.23±0.01 0.25±0.01142 MeV: hL0 .22±0 .0 1  0 . 21±0.01  0 . 21±0 .0 1  0 . 21±0 .0 1These efficiencies were obtained by averaging over each position for a given tagging dis­tance. For example, the left side tagging with the left cart at 2.0 m was run with the right side at 1.0, 1.3 and 1.5 m.The relative cross sections are calculated using both the singles and coincidence data, and demonstrate the internal consistency of the experiment. Normalizing the third counter to unity, the following preliminary cross section ratios have been obtained at T^ = 142 MeV:da j da2 da3Right singles 0.3410.01 0.52+0.02 1.00 1.3410.04Left singles 0.33±0.02 0.5410.02 1.00 1.29+0.03Coincright tagging 0.3410.02 0.5310.03 1.00 1.3210.07left tagging 0.3310.02 0.5310.02 1.00 1.28+0.05Averaging all of the positions, the resultsare given below:142 MeV neutrons 0.4410.02 0.5310.02 1.00 1.30+0.05142 MeV protons 0.3310.01 0.5410.01 1.00 1.2910.03142 MeV protons 0.33 0.54 1.00 1.33(other)a >b180 MeV neutrons 0.3110.02 0.5610.02 1.00 1.1710.05180 MeV protons 0.3710.01 0.58+0.02 1.00 1.2010.04180 MeV protons 0.36 0.58 1.00 1.21(other)a>b217 MeV neutrons 0.3910.03 0.5510.02 1.00 1.20+0.03217 MeV protons 0.38+0.01 0.5710.01 1.00 1.1217 MeV protons 0.39 0.57 1.00 1.20(other)3 .baBoswell et al., Phys. Rev. C 25_, 2540 (1982). bRichard-Serre et al., Nucl. Phys. B20, 413 (1970).10Experiment 287Measurement o f parity violation in p-pscattering (J. Birchall, W.T.H. van Oers,Manitoba; G. Roy, Alberta)During the past year efforts have concen­trated on solving some of the major technolo­gical difficulties associated with a parity violation measurement at the 10~8 level. Significant progress has been made in the following areas: 1) beam position control, 2) ionization chamber performance, and 3) design of a prototype detector for the angular dis­tribution measurement.Further beam transport calculations have been performed for the proposed upgrade to beam line IB for the parity violation experiment, and optics conditions which minimize the false asymmetry due to unwanted transverse polarization profiles have been studied. At the target the optimum beam spot is achromat­ic in x, dx/dz, y and dy/dz and has cylindri­cal symmetry, which places considerable restrictions on the configuration of elements in the beam line. Since the necessary modi­fications to beam line IB will constitute the largest expense for the experiment, alterna­tives to upgrading beam line IB have been actively considered. Beam transport calcula­tions have shown that the optimum beam char­acteristics cannot be achieved on beam line 4B due to the lack of space for installing an additional quadrupole triplet between the two superconducting solenoids in the vault area. These calculations indicate that beam line 4A is a viable option, although the lack of space in that beam line places severe con­straints on the size of the parity apparatus.Detector simulations have indicated that beam position excursions must be kept below 3 p in order to avoid generation of a false parity signal due to the coupling between unwanted transverse polarization components and beam motion. Tests of a fast feedback system based on split plate ionization chambers and air core steering magnets on beam line IB have shown that this requirement can be achieved at one location; further tests are scheduled to investigate a two-loop system controlling beam position and direction ex­cursions to the desired precision.Since last year's annual report the use of a scintillator array to measure the angular distribution of Az has been abandoned in favour of a parallel plate ionization chamber with segmented collection electrodes perpen­dicular to the beam axis. The ionizationchamber is expected to be advantageous with respect to overall stability and gain match­ing of the detector segments; however, disad­vantages include possibly greater intrinsic detector noise, e.g. due to spallation reac­tions in the electrodes and microphonic pick­up. Intrinsic detector noise was measured in a small, hydrogen-filled axial field ioniza­tion chamber consisting of two identical sense regions separated by foils of various materials. The detector noise figure a, attributed to spallation in the electrode foils, varied from 1.8 for C to 1.2 for Ta, where the counting time to achieve a desired statistical accuracy in Az is increased by the factor /(1+a^) due to the detector noise contribution. Based on these data it should be possible to measure the angular distribu­tion of Az in 6 bins from 7°-43° lab to a statistical accuracy of ±2xl0-8 in 4 weeks running time with 1 pA of longitudinally polarized protons, pz = 0.6.Based on the successful spallation noise tests a prototype angular distribution detec­tor has been designed, as shown in Fig. 8. A preliminary calculation of the detector response function is shown in Fig. 9, based on predictions of the longitudinal analysing power angular distribution by Simonius [private communication]. Upon completion of this prototype detector and scanning polarim- eters which are currently under design, work will focus on measuring and modelling the sensitivity to transverse polarization com­ponents to achieve an accurate quantitative understanding of the systematic errors, which is crucial to the success of the parity violation measurement.Experiment 297Energetic neutron spectrum from p capture in deuterons (Y.K. Lee, Johns Hopkins)Three weeks of beam time were used at TRIUMF for an experimental run during January and March. The experimental arrangement is shown in Fig. 10, with the y~ beam coming from the M20A channel. Muons stopping in the liquid deuterium target satisfied the telescope re­quirement of S0»S1*S3»S2*V, where V implies any of the veto counters VI through V7. Prompt events due to it- or e- were eliminated by vetoing any event generated within 150 ns of the beam telescope signal. A total of 3xl010 u~ was stopped at the rate of 105 y“/s.As indicated in Fig. 10, the time-of-flight (TOF) system used 8 'near' counters and 1211t  h2 outFig. 8. Prototype ionization chamber detector.'far' counters with a flight path of 2.4 m. One neutron of a back-to-back pair was de­tected in a near counter placed close to the liquid deuterium target, while the TOF of the other neutron was determined using an oppo­site far counter. The efficiency of the sys­tem for detecting back-to-back neutron pairsFig. 9. Predicted response of segmented ion­ization chamber. Solid line: prediction ofAz(6) by Simonius using 'best values' of hPP and hPP. Points: predicted signal in 10 cm segmented ion chamber, with statistical errors only, for 1 = 1 pA, pz = 0.6, 20 cm target, 700 h counting time.was 1.9xl0-it at 45 MeV for the discrimination settings used. The efficiency and energy resolution of the TOF system was calibrated using a ir~ beam. In order to reduce back­ground due to p~ decays before nuclear cap­ture, two layers of veto counters surrounded the target for charged particle rejection.In off-line analysis an angular deviation of less than 16° from the 180° back-to-back geometry was required in order to signifi­cantly reduce the background from uncorre­lated neutrons and y-rays. Such a condition on the angular resolution corresponds to low momentum neutrinos and to neutron energies above 25 MeV. For the coincidence timing resolution of 1 ns the neutron energy resolu­tion was 4.4 MeV at 45 MeV.The measured energy spectrum for neutrons above 21 MeV is shown in Fig. 11. Background due to accidental coincidences has been sub­tracted. The vertical uncertainty shown in­cludes only counting statistics, while the horizontal error bars represent the combined uncertainty due to timing and the effects of kinematic range of neutrons triggering the time-zero counters. A realistic theoretical12Neutron Energy (MeV)Fig. 10. Experimental arrangement. The |i“ telescope con­sists of scintillators SO through S4 plus V3; the charged particle veto counters consist of VI through V7 (V6, not shown is horizontal and below the target); the counters near the target provide the zero time and the far coun­ters, partially shown in the inset, provide the stop time for the TOF measurements.Fig. 11.The measured neutron spec­trum, corrected for background, is shown in (b), with the vertical uncertainties including only the counting statistics; the line is drawn to guide the eye. The theo­retical spectrum shown in (a) is from Mintz with the MEC prediction from Goulard et al. superimposed (broken line).prediction [Mintz, Phys. Rev. C 1389(1983)] reproduces the measured spectrum with a superimposed enhancement of 200% due to meson exchange currents [Goulard et al., Phys. Rev. C l b _, 1237 (1982)] over a 15 MeV range up to the end point.Thus the spectrum of energetic neutrons from M- capture in deuterons was measured for the first time, and the pion-like behaviour of the muons, predicted in this energy range on the basis of MEC theory [Goulard et al.] and PCAC [Bernabeu et al., Phys. Lett. 69B, 161 (1977)], seems to be confirmed.Experiment 301The reaction p p id  near threshold (D.F. Measday, UBC)In September we had a very successful run and now we have a considerable amount of excel­lent data. An important change in policy was to use TINA and MINA in coincidence as the ir° spectrometer (instead of TINA and a lead glass detector used in September 1985). Therewas some apprehension about using Nal detec­tors with a proton beam because we had never tried this before, the main fear being the neutron background. However, the difficulties were relatively easily overcome and the im­provement in the data was dramatic.With TINA and MINA we had to use time of flight against the rf to tag the neutron background. This worked very well apart from certain occasions when we observed proton groups at two phases in the rf cycle. This required constant monitoring.The experiment was run at proton energies of 320, 350, 400, 450 and 500 MeV. Various com­binations of angles were used: several geome­tries with the Nals at 180° to each other (40°-140°, 60°-120°, 80°-100°) and others at symmetric angles (40°-40°, 60°-60°, 70°-70°, 80°- 80°, 90°-90°).A two-dimensional plot of energy deposited in MINA against energy deposited in TINA (Fig. 12) shows a very clear band due to the ir° production. The tt° y-rays have to follow13234. O I 228. 0 I 22 2 . 0  I216 .0  I 210  0  I 204 0 I 198 0 I192.0 I186.0 I 180 0 -174.0 I 168 0 I 162. 0 I 156 0 I150.0 I 144 0 I 138 0 I 132 0 I 126 0 I 1 20 . 0  -  114 0 I 108 0  1 102 . 0  I96. 0 I 90. 0 I 84 0 I 78 0 I 72 0 I 66 0 I.60 0 -  54 0 I 48 0 I 42 0 I 36 0 I 30 0 I 2 4 .0  I . 18 0 I 12. 0 I 6  0  1 . 0 . 0 -  . ♦ I -:X*:.. ..: . *S  . .: , X # . ....................... :*##:.......: :,X#S:. : : x##x: . . . .. : : xXSix. . .. :::.X##«:...X4«...X S 4 S .............X « # i :........S * * x . ......x X X * S ....... . B X S S x : . . xXSXX.........:::::: xXxX«*Sx... .................................  ........x S X t # # S . :  .....................................• ____ x x X S * S # * S X x . :............................................................. « * x X S # # * * S # * S x . .:. ........ x* XxXS####XXXxi1. 25 * 10* *  2Fig. 12. Two-dimensional plot of energy deposited in TINA against energy deposited in MINA. The hyperbolic band is from it0 decay y-rays from p+p p+p+ir0.a hyperbola becauseMtt°e t e m  =  - — r r T T T o T  •(Et = energy in TINA; EM = energy in MINA;i)j = y-ray opening angle.) We therefore can use the mass of ir° as a way of selecting the important events.We have observed significant analysing powers (Fig. 13), and these are the first results of this observable to our knowledge.Experiment 304Muonium-antimuonium conversion (A. Olin, TRIUMFA/ictoria)The muonium-antimuonium conversion (reaction (p“*"e~ ■+ li'e"*") is expected to occur in elec— troweak models with L=2 lepton number viola­tion. Majorana masses for the neutrinos (implying L=2) reflect an attempt to solve the cosmological dark matter problem while leaving them without Dirac mass. If the masses are induced through an isovector Higgs coupling, then the doubly charged Higgs will induce muonium (Mu) to antimuonium (Mu) con­version. Stringent limits on such processes can be set if they are also flavour changing, but for intra-generation processes onlydouble beta decay imposes limits, which are quite model dependent because of the complex­ity of the process and correspond to Higgs masses on the order of 100 GeV, i.e. coup­lings of G=Gp. This experiment will be sensitive to masses in this range.7t°  Lab Energy (MeV)Fig. 13. Preliminary results of the analysing power of the reaction pp ppir0 at 500 MeV incident proton energy and 40° it0 lab angle. The results are from one run only. The errors shown are statistical. In addition there are systematic errors of about 10%.14T s T aFig. 14. Apparatus for the muonium-antimuonium conver­sion measurement using THC gas. The vacuum system is shown schematically: IP ion pump, SP sorption pump, IGionization gauge, RGA residual gas anclysol, PG piranni gauge.Fig. 15. Particle decay scheme of 181*Ta.The experimental technique consists of stop­ping y+ in Si02 powder where Mu form, diffuse into vacuum and convert to Mu. Any p- so produced will be allowed to capture on W to form 181tTa, which has an 8.7 h half-life. The resulting activity is counted off line in a low background environment.The setup is shown in Fig. 14. Muons from M15 are moderated and stop near the back of a 16 mg/cm2 thick layer of Si02 powder (7 nm particle diameter). Thermal Mu then diffuses from the powder across a 2 cm vacuum gap to a 5 nm W layer evaporated onto a Si wafer. Con­verted p~ will then capture to produce 181*Ta. Approximately 50% of the recoiling Ta atoms will remain in the foil, which may be removed from the vacuum system without disturbing the Si02 powder. The Si02 powder region is viewed by three MWPCs and MINA through a 10 cm diam­eter thin window. Muon decays originating from the vacuum region adjacent to the powder are imaged in order to monitor the vacuum yield.One then searches for the 8.7 h 184Ta isotope using a well-shielded beta-gamma coincidence array. The decay scheme (Fig. 15) shows that the 181tTa betas are accompanied by a 414 keV gamma to a metastable state (T=8 us) which subsequently decays to the ground state with emission of, on average, two gammas. Thus beta-gamma and especially beta-gamma-delayed gamma events with a 414 keV gamma are a unique low-background signature of the production of u-.A 1 week run in January was used to study vacuum Mu yields from Si02 powder targets. The results were compared with measurements on an AI  foil target (no yield expected) and with a Monte Carlo calculation, and very high yields (15-20% of the muons stopped in a3 mg/cm2 target) were observed. This measure­ment made full use of the excellent optics and luminosity of the M15 channel. The ex­perimental set-up is shown in Fig. 16. The data were analysed by fitting the MWPC co­ordinates of the decay positrons above 20 MeV to a straight track. Figure 17 shows a densi­ty plot of these trajectories extrapolated to the vertical plane containing the beam axis. Decays originating from the target and beamscintillator are clearly seen.Histograms of the observed time of decay for real events in different spatial regions are shown in Fig. 18 for a 1 mg/cm2 layer of3.5 nm powder and for the Ail foil target. These are compared to a Monte Carlo simula­tion. Note the nonexponential time dependence introduced by motion of muonium into and out of the different regions. The simulation allows estimates of partial yields for diffu­sion from the layer followed by decay in each of the four regions, per muon stopping in thesilica layer. Based on the measured yieldfor region V2 of 1.6±0.12% per stop in the silica, yields into the vacuum of 6.7%, 5.3% and 0.7% are estimated from the simulation for decays from the target, VI and V3 regions, with the remainder decaying else­where. However, the inferred total yield15Ocm./  Op/ 1r*' i1 . |JCl>- 1\ 1 \ o' - X  wMUPC1^nupc2N a I ( Tl)Fig. 16. Schematic of experimental apparatus. The powder target is indicated by P, and S is a stack of three scintillators. Inset: Target region as viewed by the wire chambers. The powder layer P, the thin scintillator S and its lucite light guide (L) are indicated, as are the co-ordinate axes. Dashed lines indi­cate structure out of the plane of the beam.depends on the input parameters and the model. From simulations with other assump­tions, in particular for isotropic emission from the layer, a systematic uncertainty of 30% and a yield of (19±6)% is indicated. The same analysis for a 1 mg/cm2 layer of 7.0 nmZ fmm)Fig. 17. Density plot of projection of decay positron trajectory to x=0 plane, showing definition of regions for time histograms, fhe co-ordinate axes are defined in Fig. 14.particles gives a similar result while a 3 mg/cm2 layer of 3.5 nm particles yields (15±5)%.A July run was devoted to measuring the yields, rates and efficiencies to establish the feasibility of the conversion measurement and to study the dependence of the yields with powder thickness to determine whether a multi-layer target is desirable. Preliminary analysis shows yields of 10% for a 13 mg/cm2 target. Backgrounds from beam-relatedFig. 18. Muon decay times for events from the spatial regions defined in Fig. 16. Filled circles represent the data from a 1 mg/cm2 powder, open squares from a 3 mg/cm2 ASL foil, and the histograms result from the simulation calculation with diffusion con­stant of 80 cm2/s.16processes [e.g. 181+W(n,p), 186W(n,t) or186W(e+,e+d)] were measured by exposing a 25 y W foil to y+ for 24 h during the run. Subsequent counting of this foil revealed a single count in the 414 keV peak, well above the continuum background rate. If 10 nm evaporated layers are used in the experiment, this corresponds to a rate of 1 count per 100 days. A data-taking run with sensitivity < Gp is planned for 1987.Experiment 332Ratio o f spin transfer coefficients(C.A. Davis, Manitoba)We have measured the ratio of Wolfenstein spin transfer coefficients Dt (sideways to sideways spin transfer) and Rt (vertical to vertical spin transfer) in quasielastic scat­tering from deuterium: d(p,n)pp at energies223, 324, 429 and 492 MeV at 9n = 9° (lab) to an accuracy of ±1%. Systematic errors are reduced by determining the ratio Dt/Rt rather than the individual spin transfer coeffici­ents, since the ratio can be determined in a manner which is independent of the analysing powers of the polarimeters. The data will be corrected for deuteron final state interac­tions to deduce the ratio T>t/Rt for free n-p scattering.The apparatus to measure D(;/®-t -*-s presented schematically in Fig. 19. The essential com­ponents consist of a liquid deuterium target (the experimental scattering target normally used as the neutron production target for np experiments in 4C), spin precession magnets, polarimeters to measure the incident proton polarization, and a polarimeter (LH2 target plus two proton 'booms' to measure trajec­tory, range and TOF of the recoil protons) to measure the polarization of the scattered neutrons. In addition there is a proton beam position feedback loop and a neutron 'beam' profile monitor.The data-taking phase of this experiment was completed this year and analysis is currently under way. The following preliminary results have been obtained at 492 MeV, where we have considered only the statistical errors in the quoted uncertainties:Dt/Rt (quasielastic) = -0.411 ± 0.009Rt (quasielastic) = -0.764 ± 0.008The values deduced for free np scattering are:Fig. 19. Schematic layout of the experimen­tal set-up. The polarized proton beam passes through the polarimeters and then through a superconducting spin precession solenoid (which is energized for Rt data and off for Dt data) and is then incident on an LD2 target. Neutrons from the d(p,n)pp reaction pass through a collimator at 9° (lab), through a vertical field spin precession mag­net, then through a horizontal field spin precession magnet (off for Dt data), and are incident on an LH2 target. Recoiling protons are scattered left or right into pro­ton range counters with full track recon­struction and TOF determination (measures the neutron polarization). A neutron beam profile monitor and four-branch polarimeter are at the end of the neutron beam line.Proton Beom4-Branch Proton PolarimeterCollimator- 9 Port'Lead CollimatorSplit Plate SEM'sProton Polarimeter (small El)Spin Precession Solenoid (L)Clearing Magnet V)Scintillator DLC'sLiquid Hydrogen TargetWedgeDegraderScintillatorsCH2 Ta4-BranchNeutronPolarimeterNeutron Beam Profile Monitor17Tn in MeVFig. 20. Comparison of the preliminary free np value for Dt/Rt to predictions of various phase-shift analyses.Dt/Rt (free np) Rt (free np)-0.457-0.725The preliminary result at 495 MeV is pre­sented in Fig. 20 in comparison with predic­tions from several current phase-shift analyses. The preliminary data are in some disagreement with current phase-shift fits. However, we prefer to complete the data analysis and investigate all systematic errors before drawing conclusions from these results.Measurement o f K + ~  i r +vvBNL 787 (BNL-Princeton-TRIUMF collaboration)(D. Bryman, TRIUMF)production of the supersymmetric partners of the photon, the graviton and the Higgs. In some versions of supersymmetry the decay KT*" > 7t+yy could occur at almost the current exper­imental limit for K+ +■ ir+vv which is 1.4xl0-7. In a recent paper Masiero et al. [Phys. Rev. Lett. J37_, 663 (1986)] show that parameters in some superstring theories are most stringent­ly constrained by K+ -*■ Tr+vv and ye con­version. Any proposed interaction which in­duces neutral flavour-changing process (e.g., technicolour) will also contribute to the K+ ■+ tt+v\T rate. Consequently, a large window for new physics exists between the current upper limit of 1.4xl0-7 and 5xl0-1®.Candidates for X° in K+ tt+X° include axions, Goldstone bosons, familons and Higgs parti­cles. The experiment will be sensitive to a wide range of lifetimes and decay modes of such particles and is expected to reach sens­itivities increased by a factor 1000 over previous attempts.Experiment 787 will also be sensitive to other interesting rare decays such as K+ + ir+YY, XT1" + ir+ye, IT1" ■» Tr+e+e~, KT1- + ir+y+y_, K+ + y+vvv, K+ > e+VY, among many others.Figure 21 shows the apparatus under construc­tion. The detector is designed to have a large geometrical acceptance (2it sr) for the K+ + 1T+ vv decay mode while maximizing therejection of background processes such asK y+v, K+ > u+vy, K~*" x IT-orO and others.Sensitivity for identification of unaccom­panied pions from K+ + ir+v\7 is accomplished through measurement of momentum, kineticThe decay K+ ir+X° (where X° stands for one or more light, neutral weakly interacting particles) will be searched for at the level <2xl0-1  ^ in Expt. 787 at the Brookhaven AGS. K+ tt+vv offers a unique testing ground for higher-order weak effects in the standard model not dominated by long-distance effects (as is the KL~Kg mass difference). For a top quark mass of 60 GeV the branching ratio is estimated to be in the range 3xl0-10 >BR(K+ ■+■ tt+ v v) > 10-10. Observation of a sig­nal in this region would serve to validate the standard model with three generations and place severe constraints on its parameters.The observation of an apparent K+ + ir+vv signal above about 5xl0~10 would be a dramat­ic indicator of new physics. The least exotic possibility is the occurrence of additional generations of neutrinos. Others include the■jysnT A B M T  A M U T  P H Q T Q h M - T P l C W SFig. 21. Experiment 787 detector1 8energy, range, decay sequence TT+y+e, and efficient rejection of both single photons and photons from tt° decay.The 800 MeV/c K+ beam is brought to rest in a 10 cm diameter target consisting of groupings of scintillating fibres 2 mm in diameter. The decay pions pass through a cylindrical drift chamber which measures their momenta in the 10 kG magnetic field with resolution a„ < 2%. The pions then stop in a plastic scintillator range stack which also contains MWPCs. Each range stack counter is viewed from both ends by 2 in. phototubes read out by transient digitizers. The digitizers record a complete history of scintillator light output as a function of time. The total energy of the decay pions will be measured by summing the pulse heights of the target and range array elements with an anticipated resolution ae ~ 3%.The detector is completely surrounded by a Pb scintillator gamma veto consisting of barrel and endcap sections. Monte Carlo calculation of photon detection inefficiencies for the proposed configuration found an overall it0 inefficiency ~5xl0-6 in the energy range of interest. The efficiency is limited by photonuclear absorption.A branching ratio sensitivity B(K+-»-Tr+vv) < 2xl0-1® will be achieved in a run of 2500 h with a IT*- stopping rate of 3xl05/pulse. De­tailed estimates based on measurements at BNL and TRIUMF and on Monte Carlo calculations indicate that <1 background event is to be expected. Limits possible on the process K+ -*■ ir+a would be in the region <10-10.The principal activities of the TRIUMF group include: central drift chamber, endcap photon veto, gallium arsenide transient recorders, beam counters and chambers, data acquisition and trigger systems, and analysis and Monte Carlo software.The magnet will be completed at BNL in January 1987 including field mapping. The detector installation will proceed during the winter and the first run is scheduled for May 1987.The central drift chamber system is being entirely developed at TRIUMF. The active volume is enclosed by Ai. endplates and graphite-epoxy cylindrical walls. The chamber is arranged in five layers of multiple sense- wire cells. Three layers are axial and two are at a stereo angle of 3.5°. The wires arestaggered by 500 ym from the cell axis to provide local resolution of the left-right ambiguity. Six central sense wires will be used resulting in up to 30 points measured on a track. The half-widths of the cells are 1 cm for layer 1 and 2 cm and 1.5 cm for layers 3, 4 and 5. There are no potentialwires adjacent to or between the sense wires. Crosstalk between sense wires after compen­sation is <5%. The Lorentz angle of drift is about 25° for a magnetic field of 1 T. There is high efficiency and a good distance versus time relationship is maintained throughout the cell.The wires are tensioned between 0.95 cm thick A l endplates spaced by 50 cm between 0.51 mm thick cylinders. The inner cylinder will be graphite rigidly attached to the end plates. The outer cylinder is installed with an 0-ring gas seal after the chamber wires have been strung. The end plates are prestressed according to calculations to maintain uniform wire tension. The sense wires are positioned by precision grooves in a molded Reitan feed- through. Each injection molded feedthrough is epoxied in a precision-machined slot in the endcap and carries a finger stock printed circuit board for gluing and soldering the tensioned wires. End plate machining was completed in October.The electronics for each sense wire was de­signed at TRIUMF and assembled in local area commercial electronics shops. It consists of an on-board low power (30 mW) hybrid preamp­lifier with output transmitted to an ampli­fier discriminator at a distance of approxi­mately 35 m from the chamber. Assembly of the preamp and postamp hybrids began in December. The TDCs purchased were the LRS 1879, 500 MHz Pipeline TDC. The measured rms resolution from the electronics system is approximately ot ~ 1 ns, resulting in a50 ym uncertainty for a typical gas with drift velocity v = 5 cm/ys.Single cell and multiple cell prototype cham­bers were constructed and tested extensively under realistic conditions including magnetic fields. Local position resolution of a ~ 150 ym was obtained. Track fitting of data from 200 MeV/c pions from the Mil beam at TRIUMF gave position resolution consistent with <2% momentum resolution for the 787 chamber.The TRIUMF group is also building the two lead scintillator endcap veto detectors. A petal geometry was chosen where the NE10419scintillator and lead sheets are positioned transverse to the beam axis and the 24 azi­muthal segments are read out via BBOT wave­length shifter bars by phototubes located outside the magnet. Bench tests conducted over the spring/summer period were confirmed in beam tests at TRIUMF. The detector will produce 10 photoelectrons per visible MeV energy deposited which guarantees that a low threshold of 1 MeV is attainable. The rise time of the pulses is of order 5 ns and the full width at 1/10 maximum is of order 30 ns. The two endcaps are scheduled for completion in March and May 1987.A potentially low cost (<$200/channel) trans­ient recorder based on a gallium-arsenide charge-coupled device (CCD) is under develop­ment at TRIUMF. Prototype CCD devices have been produced and are undergoing measure­ments. The devices have 64 charge buckets which would allow storage for 128 ns at 500 MHz. It is intended eventually to produce a 256 bucket device. Initial devices have exhibited charge transport over a frequency range of 1 to 500 MHz. During the past year a new microstructure laboratory for production of the CCDs was built at TRIUMF. First devices from this lab (32 and 64 bucket CCDs) are under construction. Development of ancil­lary logic, supervisory and FASTBUS digital readout are also in progress at TRIUMF.The beam counter system consists of several scintillators, two scintillator hodoscopes, three planes of MWPCs and a Cerenkov counter. All but the latter are being built at TRIUMF. The MWPCs will use a fast CF^ gas mixture and special home-built hybrid electronics includ­ing the postamplifier/discriminator system built for the central drift chamber.For the data acquisition task the 787 group has chosen a multiprocessor system using the microVAX II. For reasons of speed and high density the majority of instrumentation in the experiment is FASTBUS based. Provisions are made for a CAMAC—FASTBUS interface for the CAMAC modules needed. The SLAC scanner processor (SSP) is the FASTBUS crate control­ler which provides enough power and flexibil­ity to execute energy triggers at the crate level and data transfers in the interspill time functioning as a segment interconnect. During the past year the TRIUMF group commis­sioned a two-SSP data acquisition system at BNL. A system under development at TRIUMF for the drift chamber is also operational. The trigger system consists of three levelsarranged in order of successively increasing complexity and deadtirae.All three collaborating institutions work primarily on VAX/VMS systems. Histogramming is done with the TRIUMF FIOWA/REPLAY package along with the TRIUMF graphics software. Data tape and file handling has been via the TRIUMF package MAGTASS/MAGTA/10. Monte Carlo programs have been developed at Princeton and TRIUMF. The CDF YBOS data acquisition system has been adopted by the 787 group. Work is already under way on reconstruction and moni­toring routines for the various subsystems (target, drift chamber, range stack, barrel veto and endcaps).The SLD group(A. Astbury, TRIUMFA/ictoria)SLD will be the major new detector used to explore Z° physics at the Stanford Linear Collider (SLC), and physicists from TRIUMF/ UBC/Victoria are involved in the collabora­tion. SLD has been designed to include all of the features believed to be essential to exploit fully the physics potential offered by the projected high rate of Z° production. In particular, it will be able to search for and study rare processes, e.g., Higgs produc­tion and supersymmetric particle production. The detector has good vertex detection and charged particle tracking, plus hadron and lepton identification over essentially a 4ir solid angle. The lepton identification also includes neutrinos since SLD, unlike the MARK II which will explore the early SLC physics, has full solid angle coverage of electromag­netic and hadron calorimetry.The Canadian group based at TRIUMF has been responsible for the specification and de­tailed design of the barrel liquid argon calorimeter (LAC) (~1.48° 7 0 ? ~32°). The radiator will be lead, 2 mm thick in theelectromagnetic section and 6 mm thick in the hadronic section. Charge will be collected from the 2.75 mm liquid argon gaps. The pro­jective tower geometry is derived from aseries of lead tiles held in place by plasticspacers and alternating with continuous lead sheets. Figure 22 shows the principle em­ployed in the construction. The anticipated resolutions are ~8%//E for electromagnetic (EM) energy and ~60%//E for hadrons. In order to prove the mechanical design a suc­cessful prototype has been constructed at TRIUMF. The high voltage tests are also20Fig. 22. Exploded view of partial tile array.satisfactory, with <1 nA leakage current per tower, measured in dry air under typical operating voltages ~2-3 kV. The structure will be immersed in liquid argon and tested as a calorimeter at SLAC in January 1987.The meson hall extension is being converted into a production facility, and 75 EM modules will be manufactured commencing in February 1987, with completion envisaged for early in 1988 when installation into SLD at SLAC will commence.The group also has a major responsibility for the provision of the off-line software asso­ciated with the LAC. The work entails Monte Carlo simulation of the calorimetry as well as the preparation of analysis programs. The main code is produced on the IBM at SLAC, where some members of the group are in full­time residence. In addition, MC simulation studies are currently pursued on the UVic and UBC campus computers.21NUCLEAR PHYSICS AND CHEMISTRYExperiment 208 Proton-proton bremsstrahlung  (P. Kitching, Alberta/TRIUMF)The analysis of this experiment, which com­pleted data-taking in 1985, is nearing com­pletion. Last year's annual report described the object of the experiment, which was to look for off-shell effects in the nucleon- nucleon force, and the experimental apparatus and technique. The results of the analysing power measurements will now be discussed. The cross sections are still being analysed.The experimental results were compared with theoretical calculations of two different types, namely the soft photon approximation (SPA) and potential model calculations. In the soft photon approximation the cross sec­tion for the ppy process may be written as ^-2--- = —  + 2ReAB* + (B2+2ReAC*)k + 0(k2)dflgd^dBy kwhere k is the photon momentum and A, B and C are coefficients arising from the expansion of the ppY amplitude about the on-shell (k=0) point. The A and B coefficients contain kine­matic factors and purely on-shell (elastic) information, while C and the higher-order coefficients contain off-shell information as well as higher-order on-shell contributions. The soft photon approximation as used here assumes that C and 0(k2) terms are negligible but includes the B2 term of 0(k). New poten­tial model calculations were carried out using the Paris and Bonn potentials, which are generally believed to be the best avail­able at the present time. Under the kinematic conditions applicable to this experiment re­sults obtained with Paris and Bonn potentials differed very little, the differences in the calculated analysing power for instance being always less than 0.07.When these calculations are compared with the measured values of the analysing power the following conclusions may be drawn:1. At the larger proton opening angles where the photon energy is small and the NN amplitude is nearly on shell the measured analysing powers are consistent with zero and in agreement with the (small) pre­dicted analysing powers whether calculated in the SPA or either potential model.2. At the smaller proton opening angles, the measured analysing powers are in strong disagreement with the SPA predictions, but in good agreement with both the Bonn and Paris potential models. Overall, the value of x2 per degree of freedom is ~1.5 for both Paris and Bonn potentials and is -5.5 for the SPA. Figure 23 shows the angular distribution of the analysing power for some of the twenty angle pairs at which data were obtained in the experiment.Thus, in contrast to earlier ppy experiments the measurements reported here are in strong disagreement with the predictions of the softphoton approximation, which incorporates onlyon-shell amplitudes. The data are in goodagreement with potential models using either the Bonn or Paris potentials, which differ little in their predictions for the kinemat­ics of interest here. This conclusion may be interpreted to mean that the Bonn and Paris potentials differ little in their behaviour for off-shell momenta probed by this experi­ment, estimated from the kinematics to be -1-2.5 fm"1. This statement should be quali­fied by noting that at 280 MeV the experiment is more sensitive to the P-wave amplitude, whereas calculations show that the Bonn andParis potentials differ most in the S-wave off-shell behaviour. However, at lower ener­gies where S wave would be more important, off-shell effects will in general be smaller.In any case the experiment is the first direct measurement of the off-shell behaviour of the N-N force, and the fact that it is in agreement with modern potential models of the N-N force increases our confidence in the essential validity of these models [Kitching et al., Phys. Rev. Lett, (in press)].Experiment 234Studies o f simple features o f the A(p,x~)A + 1 reaction in the 3,3 resonance region (G.J. Lolos, Regina; R. Bent, IUCF)To elucidate the role of the A isobar in the (p, tv ) reaction mechanism the authors have performed an experiment using unpolarized beam to study the energy dependence of the (p,ir+) and (p,ir~) reactions on 13C leading to mirror states in and ^0. By choosingmirror final states effects due to diffe­rences in nuclear structure are minimized,220 .5 00.00 —HER  L E P  1 2 .4  14-.O S P A B O N N PA R ISV h(a)3 0  6 0  9 0  120 150 ISOR H O T O N  ANGL.EIH ER L E P  12.4- 2 2 . 0 SR A B O N N P A R IS(b )3 0  6 0  9 0  120 150 ISOR H O T O N  A N G L E— .25  —HER L E R  1 7 .3  2 2 . 0 S R A B O N N PAR IS(c)3 0  6 0  9 0  120 150 180R H O T O N  A N G L EHER L E R  ---------S R A2 7 . 8  2 2 . 0  ---------B O N N R A R 1Is,T i ^ r u * ? n(d)O 3 0  6 0  9 0  120 150 180R H O T O N  A N G L EHER L E R  ---------SR A2 7 . B 2 8 . 0  ---------B O N N P AR ISI(e)O 3 0  6 0  9 0  120 150 ISOP H O T O N  A N G L EFig. 23. Measured analysing power as a func­tion of photon angle, together with theoreti­cal predictions. HEP and LEP are mean angles of high-energy proton and low-energy proton, respectively.and any differences in the energy behaviour of the (p ,tt+) and (p,n_) differential cross sections at the same momentum transfer can be attributed to the reaction mechanism. For added comparison the energy dependence of the 12C(p ,tt+) 13C reaction to similar final states was also studied.This experiment received time on the MRS spectrometer in January 1985 (9x12 h shifts) and July 1985 (10x12 h shifts) and is nowcompleted. Data analysis was also completed in Regina using the Lisa Analysis System. A sample off-line analysed pion spectrum from the first run (after solid angle, energy loss, time of flight and target illumination cuts) is shown in Fig. 24.Because (p,ir_) reactions have very low cross section, background problems warrant careful consideration. Fortunately, part of the back­ground problem associated with (p,ir+) experi­ments is absent in (p,ir-) experiments due to the negative charge of the pions in the mag­netic spectrometer; this coupled with the redundancy of the cuts applied in the off­line event-by-event analysis made it possible to observe states with differential cross section as low as 0.5 nb/sr. The energy reso­lution obtained in the first data-taking run was typically 180 keV at 354 MeV and in the second run the resolution was approximately 300 keV at the same energy. The energy reso­lution of the second run was due to a larger than usual energy spread of the incident proton beam from the TRIUMF cyclotron.2 0Oo5 -C(PlTT-) 0 Tp = 250 MeVe,„*29*ill T mi i |i mi l ip7- 1 1  3 5  9 11E x c i t a t i o n  E n e r g y  ( M e V )Fig. 24. Expanded section of a (p,ir“) focal plane spectrum from the first run of the experiment.23t  =  0 .5 0  GeV2 / c 21200- —.....  - --1-''9 006003 0 0 -0*0oA o□<?.(35 1X 5 1.251 2 C ( p , n + ) 13 C ? .5  1.XX 1.456 4 0  ' 4 8 0  ' 3 2 0  : 1601.05 4 0 J 3 0 - 20- 10-? 0 51.15 1.25i“ctp^ +j'-C t . i1.35 1.451.15 .1.451.25 ' 1.35/ s  -  (G e V )Fig. 25. Differential cross section versus centre-of—mass energy s at a constant four- momentum transfer t of 0.50 GeV2/c2 . Plotting symbols indicate the source of data as follows: • This work; □ Dahlgren et al.;#  Soga et al.; O  Vigdor et al.; o Lolos et al. The error bars reflect statistical uncertainties only.Fortunately, this reduced energy resolution did not affect the minimum observable cross section. A plot of differential cross section versus energy for the13C(p,TT+)11+C*6 3~,13C(p,^-)llt0*6>2)3- and 12C(p,^+)13C*g '5 ^ /?+ reactions was compiled with data from this work as well as other previously published data and is shown in Fig. 25. The data are displayed using the relativistically invari­ant Mandelstam variables where s is the square of the centre-of-mass energy, and t is the square of the four-momentum transfer. Cross sections at a constant value of t = 0.50 GeV2/c2 were obtained from measured data using either an exponential or Legendre polynomial fit.Unfortunately, the small (p,t ) cross sec­tion, coupled with a 2.1 msr spectrometer solid angle and an 11 m spectrometer path length, made it impractical to obtain a dis­tribution of (p,tt~) data over a large angular range (corresponding to a large range of t values). Nevertheless, the smooth exponen­tially decreasing angular distributions that(p,n) reactions are known to exhibit above 250 MeV lead us to expect that the s depen­dences displayed are not too sensitive to the t value chosen.Figure 25 indicates that (p,ir+) reactions have an enhancement in differential cross section which is maximum at the invariant mass of the A(1232). This enhancement has been noted twice before by Couvert and Dillig and also by Hoistad. In contrast, the diffe­rential cross section of the (p,ir-) reaction shows no enhancement near the invariant mass of the A(1232); these are the first measure­ments of the energy dependence of the (p,ir-) reaction in this dynamical region. The lack of any enhancement of cross section in the A region was also seen in the 13C(p,u-)140*g.s. reaction which is not presented here. That this is a common feature of two transitions to two entirely different final states indi­cates that there is a substantial dynamical difference between (p,ir+) and (p ,tt~) reac­tions which is not dependent upon nuclear structure.Experiments 238, 252Validity o f DWBA analysis for intermediate energy proton scattering to low-lying states of 20ePb (ORNUOregon/Oregon State)New measurements of inelastic proton scatter­ing to low-lying collective states of 208Pb have been made with incident protons of 200 and 400 MeV. Angular distributions around the first diffraction peak have been measured for the states at 2.614 MeV (3 ), 3.198 MeV (51), 3.709 MeV (5,), 4.086 MeV (2 ) and 4.324 MeV (4+). Deformation lengths have been extracted with the collective DWBA code ECIS79 and com­pared with results from other (p,p') experi­ments .Our deduced values of deformation lengths (Table I) obtained from inelastic proton scattering measurements to low-lying states in 208Pb which are strongly excited by inci­dent protons in the 200-400 MeV range are in reasonable accord with other values. The present results for the 3~, 2+ and 4+ states, along with those measured at other energies, are plotted in Fig. 26. In general the indi­vidual values of the measured deformation lengths for each state are found to be the same, that is, independent of the incident proton energy. The uncertainty in this con­clusion is at about the ±6% level. We would argue that this conclusion is valid as long as the transition density is surface peaked24Table I. Comparison of Mjj/Mp from present analysis to results obtained from other proton scattering measurements.State <«H>3 Mn /Mp(200-400)Sn/Mp(low energy) (800)3~ 0.788±0.018 1.0410.04 1.06+0.04 1.1210.032+ 0.403+0.021 1.0010.08 1.0110.03 1.2810.054+ 0.50010.020 0.9510.07 0.9310.12 1.10+0.065T 0.35010.023 0.9310.11 1.0310.08 1.1810.070.22510.021 0.72+0.14 0.8910.06 1.10+0.07aAverage value calculated from the 200, 201, 334 and 400 MeV results.and the transition is collective. The above results showing that deformation lengths are constant over an incident proton energy range up to 800 MeV support the conclusion that the collective DWBA analysis is valid over this energy region.An important application of 5jj measurements is ^to calculate multipole moment ratios, Mn/Mp. When the DWBA analysis is reliable, we can extract deformation lengths from measure­ments at different incident proton energies and calculate the associated Mn/Mp ratios from the relation,6p (N /Z )(bPn/bPp) Jwhere is the measured hadronic deforma­tion length, 6p is the^ corresponding elec­tromagnetic value and Mn/Mp = (Mn/Mp)(Z/N). We then compute the average value and uncer­tainty following standard statistical proce­dures. The results for Mn/Mp from the present analysis are tabulated in column 3 of Table I, where the correction for different neutron and proton matter densities has been made. Within the uncertainty the values for the 3", 2+ , 4+ and 5" states are equal tounity and are completely consistent with iso­scalar excitations. Only the value for the 5~ state differs significantly from unity.Within uncertainties the Mn/Mp values obtained by us from an analysis of the data in the 200-400 MeV energy range (column 4, Table I) agree with those calculated using the defor­mation lengths deduced at lower energies. On the other hand, the values found from the 800 MeV measurements are higher than both ourFig. 26. Plot of deformation length as a function of incident proton energy. The shaded band indicates our estimate of the un­certainty involved in the computed average from an analysis of the results obtained in the 200 to 400 MeV range.200-400 MeV results and the averaged low en­ergy ones. Since Mn/Mp is a property of the target nucleus it cannot vary with energy. The difference cannot be due to the wrong choice of the neutron to proton force ratio bPn/bPp as the results are not sensitive to this parameter. One explanation is that the absolute normalizations for the 800 MeV data are slightly off. This is unlikely, however, as this experiment was done very carefully. Another possibility is that the reaction mechanism is beginning to fail slightly at 800 MeV. Our view is that the most likely explanation concerns the neglect of the spin-orbit term in the optical model used for the analysis of the 800 MeV data. The ratio of our values to the 800 MeV ones is 0.88±0.06. The authors of the 800 MeV study also measured deformation parameters for six low-lying states of 90Zr and analysed them in the same fashion as for 208Pb. A similar experiment was done at 800 MeV on 98Zr by another group and analysed with a DWBA code including the deformed spin-orbit term. The average ratio of 6jj measured in this latter experiment to the corresponding values measured in the earlier 800 MeV ex­periment is 0.83±0.03. We suspect that these differences are real and point out the need for including the spin-orbit optical model term in the DWBA analysis of all experiments.25Experiment 266Initial studies o f the (n,p) reaction on lightnuclei (K.P. Jackson, TRIUMF)The TRIUMF (n,p) facility utilizing the MRS was installed for the second time during the March shutdown. Following a brief period devoted to re-establishing full operating conditions, seven shifts were allocated to the completion of Expt. 266. In the studies of the (n,p) reaction on 8L1, 12C and 13C at 200 MeV the time was used first to confirm the measurements made in December 1985 of the cross sections at 0° and then to extend the angular distributions to 15° and 20°. Limited data were also obtained for the 7Li(n,p)7He reaction at 0°, 8° and 20°.An important goal of the measurements at 0 is the precise calibration of the (n,p) reac­tion as a probe of Gamow-Teller B+ strength, the nuclear response to the one-body operator ot+ . The measured cross sections, suitably corrected for finite momentum transfer, are compared to the strengths derived from the B" decay of 7He, 12B and 13B. The measured ratios (in mb/sr-unit strength) are 10.1, 9.6 and 10.1 for A = 7, 12 and 13, respectively. Final analysis may permit some reduction in the estimated uncertainty of ±7% for each of these values. These three results are con­sistent with detailed predictions based on the distorted wave impulse approximation and the first two with analogous (p,n) measure­ments [Taddeucci et al., Nucl. Phys. A (in press)]. In the case of the ^3C(p,n)^3N reaction, however, Taddeucci et al.'s mea­sured ratio is significantly larger (~45%) than both prediction and the corresponding TRIUMF (n,p) result.Experiment 268The 208Pb(n,p) reaction at 200 MeV(S. Yen, TRIUMF)The original motivation for this experiment was to search for an enhancement of 1+ states excited via the 208Pb(n,p) reaction if there are significant A-hole components in the wave functions of these states [Brown et al., Phys. Rev. Lett. 50, 658 (1983)]. Five shifts of beam time in May were used for a feasibil­ity study. Data were taken at 0°, 3°, 6°, 9°, 13° and 17°. A 0° spectrum, with instru­mental background subtracted, is shown in Fig. 27. The events to the left of Qnp = -4.2 MeV are due to imperfect subtraction of the instrumental hydrogen hydrogen peak and should be ignored. The broad, prominent bump- G U p (fte-V)Fig. 27. 208Pb(n,p) spectrum at 200 Qnp = -22 MeV of a ~ 4 mb/sr is consistent with the spin isovector monopole giant reso­nance in width, excitation, cross section and angular distribution [Auerbach et al., Phys. Rev. C 30, 1032 (1984); Klein et al., Phys. Rev. C 11^ , 710 (1985)]. This peak is alsoclearly visible at 3°, 6° and 9°. If this initial identification is correct this would be the first unambiguous observation of this mode of nuclear vibration. The sharp peak at Qnp = -10 MeV is tentatively identified at spin dipole [Krmpotic et al., Nucl. Phys. A342, 497 (1980)]. The 0° minus 3° difference spectrum shown in Fig. 28 should show a posi­tive signal in the region of GT excitations; a positive signal is indeed seen in the region Qnp = -10 MeV to -26 MeV, and what ap- appears to be a sharp peak is visible at Qnp = -16 MeV. More data with better energy resolution are required to firmly establish the apparent GT signal.Experiment 272Gamow-Teller strength in 24M g from spin-flip cross sections in inelastic proton scattering (O. Hausser, SFU/TRIUMF)Spin-isospin excitations provide the strong­est and best understood features in inelastic proton scattering spectra at extreme forward angles. This is seen in the momentum spectrum of Fig. 29 which was obtained with the medium resolution spectrometer (MRS) in the disper­sion-matched mode for the 21+Mg(p,p') reaction26- C U p (KeV)Fig. 28. spectra.Difference of 0° and 3° 208Pb(n,p) GT strength should exhibit a posi­tive 250 MeV and 9iab = 2.9°. Typical MRS small angle (p,p') spectra are characterized by low instrumental background and a resolution of ~140 keV FWHM. The strong 1+ peak at Ex = 10.71 MeV is a pure AT=1 isovector excitation whose cross section should be proportional to the Gamow-Tellow strength B(GT) for this particular state.The natural parity excitations (AS=0) appar­ent in the unpolarized proton spectra can be completely eliminated when the spin-flipprobability Snn is measured with polarizedprotons. This is shown in Fig. 30 where dif­ferential cross sections a and spin-flipcross sections oSnn are compared at Slab = 2.9°. Up to 40 MeV in excitation were exam­ined using the focal-plane polarimeter of the MRS at a single dipole-field setting.The spin-flip cross section crSnn is a direct measure of AS=1 spin transfer strength in 2l+Mg. At small angles (2.9°) and small exci­tation energies (Ex < 15 MeV) most of the oSnn strength arises from 1+ ,AT=1 Gamow- Teller excitations with contributions from isoscalar (1+,AT=0) and higher multipole (AL>0) excitations amounting to 20-30% of the total strength. We show here that Gamow- Teller strength extracted from aSnn in (p,p') agrees with existing data on (p,n) and (n,p) charge exchange reactions in this mass region.Ex (M eV)Fig. 29. Momentum spectrum of 250 MeV pro­tons inelastically scattered from 21+Mg at 01ab =2.9 .S.blc•ereEx (MeV)4.0 3.5 > 3.0 \  2.5*4DO\  2.0 eOJ .  1.5g 1.0COb °-5 0.0- .51 ■ ■■ 1 I■p.1 1 1 1 1Mg(p.p) 250 MeV 0.1 MeV Bins 2.9 deg., r ,i i  i i i i i i10 15 20 25 30 35 40Ex (MeV)Fig. 30. Differential cross section (top) and spin-flip cross section (bottom) for 21tMg(p,p') at 250 MeV and 0iau = 2.9°.Gamow-Teller strength was extracted from these data by calculating a and Snn in the framework of the DWIA using the computer code DW81. These calculations have the following ingredients:270 cu(deg) ®ck(des)Fig. 31. Differential cross sections (left) and analysing powers (right) for 24Mg(p,p) elastic scattering at 250 MeV. The dashed lines represent calcula­tions using a nonrelativistic phenomenological optical model.101a■db"d2 4 M g(p,p ') 200 MeV10.71 MeV T=1 f(LF ,W B)=0.86____i_____ i_____ i----- 1—6eo(d e g )12 1510‘>  10 1§V. 10*b1<T101a•db■d10108 (deg)6e9(deg)121 1 24M g(p .p ')T  1250 MeV10.71 MeV T=1 f(LF ,W B)=0.78 \• 1 1 1o GO <2> ®P9 12 15 (d eg )i i24M g(p ,p ') 250 MeV10.71 MeV T=1 f(LF,W C)=0.87i i — i-------------------- 1-------------15Fig. 32. Angular distributions for excitation of the 10.71 1+ ,T=1 state in 2'+Mg(p,pI) at 200 MeV (left) and 250 MeV (right). Transition densities are from Wildenthal and Brown (top) and from Wildenthal and Chung (bottom).281) Distortion effects which reduce the 1+ cross sections by about a factor of two were estimated from the elastic scattering data shown in Fig. 31. Phenomenological optical potentials derived from these data [Lin, MSc. thesis, SFU, 1986] imply reaction cross sec­tions or in agreement with experiment.2) The effective interaction of Franey and Love [Phys. Rev. C 3^, 488 (1985)] was used. Isovector 1+ excitations in 28Si between 200 and 400 MeV are well described by this inter­action (see Expt. 335). To relate our results to those from charge exchange reactions we calculated the (p,n) cross section per unit B(GT) at 0 = 0°, q = 0 to be 7.05 mb/sr at 200 MeV and 6.55 mb/sr at 250 MeV.3) Transition densities of Wildenthal and Brown [Wildenthal, in Prog. Part. Nucl. Phys. 11 (Pergamon, Oxford, 1984), p.5; Brown, private communication] and of Wildenthal and Chung [Chung, Ph.D thesis, Michigan State University, 1976] were used. Both of these are in qualitative agreement with the distri­bution of spin-flip strength in Fig. 30.Angular distributions for the strong 1+ ,T=1 state at 10.71 MeV are shown in Fig. 32. The 200 MeV results were obtained at the MRS in collaboration with C. Djalali, A. Galonsky and G. Crawley from Michigan State University. The theoretical angular distributions are multiplied by 'quenching factors' f = oeXp/ atheory t0 obtain agreement with experiment. In Table II the quenching factors for the hadronic reaction f(p,p') are compared to those for the electromagnetic reaction f(e,e') [Titze, Z. Phys. 220, 66 (1969);Johnston and Drake, J. Phys. A7_, 898 (1974)]. The latter are calculated with effective g factors which contain effects of meson- exchange currents and higher-order configura-Table II. Quenching of Gamow-Teller and Ml strength in 24Mg.Energy(MeV)Measuredquantityf(p,p')a WB WCf(e,e')b WB WC9-15 aSnn 0.78(8) 0.84(8)10.71 a 0.82(8) 0.90(8)10.71 B(M1) 1.18(15) 1.13(15)aTransition densities from Wildenthal and Brown and from Wildenthal and Chung, and for Opn/B(GI) = 7.05 at 200 MeV, 6.55 at 250 MeV.^Calculated with effective g factors which fit a large body of Ml data in the (s,d) shell. The B(M1) = 3.1±0.4 is from Titze; Johnston and Drake.tion mixing, i.e. f(e,e') = 1 would be ex­pected for ideal transition densities. The fact that f(p,p')/f(e,e') ~ 0.75 implies that the effective axial vector coupling constant«A < 4 ree lnThe spin-flip cross section at 250 MeV and 2.9° between Ex = 9-15 MeV is (aSnn)eXp » 2.50 mb/sr. From differences in the (<JSnn) spectra at 2.9° and 6.6° we estimate an L=0 fraction of 0.85; thus (oSnn,l+ )eXp“2.13 mb/ sr. The quenching factors for the combined isovector and isoscalar 1+ states f(oSnn,l+) are in good agreement with those for the 10.71 MeV 1+ ,T=1 state.In summary, about 80% of the Gamow-Teller strength predicted by the shell model with §A = ^A166 f°unc* i-n che (P»P') experiment. This fraction is much larger than found pre­viously [Anantaraman et al., Phys. Rev. Lett. 52, 1409 (1984)] in 28Si(p,p'), but onlyslightly larger compared with Gamow-Teller strength observed in the 26Mg(p,n) [Madey et al., Kent State preprint, 1986] and the 26Mg(n,p) [Alford et al., Expt. 357] charge exchange reactions, provided the same value for Opn(0°, q=0) (i.e. 7.05 mb/sr at 200 MeV) is used. Identification of the missing L=0 strength above Ex = 15 MeV is hampered by the large L>0 spin-flip strength.Experiment 278Inelastic scattering o f 30 and 50 MeV 7r+ pro­jectiles from the 01 state in 12C (T.E. Drake, Toronto)Differential inelastic cross sections for *2C(tt,tt') at an incident energy of 50 MeV have been measured from 30° to 130° (lab). Analysis from these measurements has been completed. The results for the -*■ 02 mono­pole transition provide evidence for the EELL (Ericson-Ericson Lorentz-Lorenz) effect and the suppression of multiple scattering in the pion-nucleon system. A full account of the experimental results is given in our publica­tion [Lee et al., Phys. Lett. 174B, 147(1986)] .Experiment 281Investigations o f the pion absorption reaction 6Li, 12C( t t±, X 1X 2) A  (G.J. Lolos, Regina)This experiment received 12x12 h shifts of Mil beam time in the December 1985 and Janu­ary 1986 running period. The detection system was the same as described in the 1985 annual report and is shown in Fig. 33; the detectors29Fig. 33. Schematic diagram of the telescopes- hodoscope arrangement.are all of NE102A plastic scintillator mate­rial and the multi-wire proportional chambers were standard 20 cm and 30 cm TRIUMF chambers. The system has the capability of particle identification and total momentum informa­tion, thus allowing the extraction of missing (recoil) momentum measurements. The beam flux was determined by two independent methods, one using two small counters in coincidence intercepting the beam halo (mostly muons) and the other incorporating an in-beam grid of scintillator strips that provided in addition to the absolute flux a beam profile with 1 cm2 resolution.A limited angular distribution was obtained at T + = 100 MeV from the 6Li(ir+ ,X1X2)A reac­tion while some diagnostic data were also obtained at T^ — 200 MeV for both 8Li and 12C targets. The system was checked and cali­brated using the ir^ d •+■ pp reaction. The in­strumental energy resolution is approximately 5 MeV for an incident proton energy of 120 MeV, thus allowing us to extract meaning­ful missing momentum spectra, as shown in Fig. 34. In this figure the transitions to the ltHe(g.s.) and the low-lying excited states are clearly shown, thus indicating strong contribution from the ir++(pn) + pp channel; indeed the majority of n6Li > pp(?) events are concentrated in the missing (recoil) momentum range below 200 MeV/c whereas the momentum of a third (unobserved) nucleon sharing the incident pion four-momen­tum would be in the region of 400 MeV/c.Based on the proven capabilities of the Expt. 281 apparatus, the TRIUMF EEC approved 33 x 12 h shifts for the study of pion ab­sorption on 6Li and C as a new experiment.Experiments 283/295Elastic and inelastic proton scattering from 20SPb at medium energies(T.E. Drake, Toronto; C.A. Miller, TRIUMF)Elastic and inelastic cross section and analysing power angular distributions for 208Pb(p,p') at incident proton energies of 200 and 400 MeV are reported (Fig. 35). The measurements range from 6° to 90° for the 200 MeV data and from 3° to 60° for the 400 MeV data. The importance of including exchange and/or medium effects in the micro­scopic models (both relativistic and nonrela- tivistic) was already demonstrated by our 200 MeV elastic data [Miller et al., Phys. Lett. 169B, 166 (1986); Lee, Drake et al.,Eastern Regional Subatomic Physics Conference March 1985, Montibello]. This inelastic data will provide further tests of the various microscopic calculations. In addition, the effect of the nuclear medium can be examined more closely since the 3“ transition samples deeper into the nucleus. Theoretical calcu­lations are under way.Fig. 34. Missing momentum for the 6Li(ir+ ,pp) 4He reaction at T^ = 100 MeV. The value of 100 MeV/c has been added to avoid negative numbers.30L A B  A N G L E  LAB ANGLE  (D E G R E E S )Fig. 35. Cross section (a) and analysing power (b) angular distribution of the 3- state in 288Pb at an incident energy of 200 MeV.Experiments 322, 394Measurement o f ir± differential cross sectionsat Tw = 90 MeV (D.R. Gill, TRIUMF; J.J. Kraushaar,R. Ristinen, Colorado)Recently Frank et al. [Phys. Rev. D 28^ 1569 (1983)] have measured positive and negative pion proton scattering cross sections at energies from 30 to 90 MeV. Their results showed substantial differences from previous data at energies below 100 MeV. These cross sections are required for the determination of the pion-nucleon phase shifts which in turn are essential for understanding the results of pion nucleus experiments and for understanding the basic pion-nucleon inter­action itself.Experiment 322 was therefore undertaken in an attempt to clarify this situation. It was carried out with simple apparatus in order to avoid as many sources of systematic errors as possible. Data were taken at seven energies for positive and negative incident pions between 66.8 and 138.8 MeV. The results have now been published [Brack et al., Phys. Rev. C 34, 1771 (1986)].The discrepancies between the Frank et al. results and older data extend to energies below those measured in Expt. 322. As well, if the Frank et al. data are not accepted there is a paucity of data, particularly in the ir~p scattering case in the energy regime 20 to 60 MeV. Thus an experiment has been proposed (394) that addresses the need for clarification of the cross-section discrep­ancies as well as the need for more data.Experiment 394 will be carried out using the apparatus shown in Fig. 36 where the target T is a Cerenkov scintillator in which the recoiling protons are detected. A test was made of this technique in six shifts of very low beam intensity during the polarized beam period in September. These test runs, made on M13 with 60 MeV positive pions, success­fully demonstrated the feasibility of the active target method. Figure 37 (a,b) are dot plots of the target ADC versus the time of flight of the particle detected by pion arms at 60° and 100°, respectively. More than 50% of the continuum events are due to the 12C(ir,2p) reaction and can be removed by applying the V condition.31Fig. 36. Arrangement of scintillation coun­ters for simultaneously measuring irp elastic cross sections at six pion scattering angles. For the angles shown at T^ = 60 MeV, and for a 400 mg/cm2 target scintillator, fewer than 8% of the recoil protons will escape from the target. Event definition is S1*T* tt1«tt2 with software cuts on pulse amplitude in T and on (T-tt2) TOF. The veto counter V eliminates 12C(ir,2p) events.Experiment 323Giant resonance study with 75 M eV it*(D.R. Gill, TRIUMF)Previous studies of the production of giant resonances in nuclei with pions have been carried out at energies near the A resonance. Some very interesting effects due to the selectivity of positive (negative) pions for protons (neutrons) have been seen [see, for example, Seestrom-Morris et al., Phys. Rev. C 33, 1847 (1986)]. Since this selectivity is even higher for large angle scattering of 50 MeV pions (Fig. 38) an experiment to look for this effect has been carried out with the QQD spectrometer on the M13 channel.Experiment 323 received 11 shifts of beam in December. Experimental data were taken at 50 MeV for both positive and negative pions scattered from ‘t0Ca over the angular range from 70° to 120° in the lab. The data have not yet been analysed, but initial results should be available early in the new year.360 *4 0Pi2 up+dn TDC—i—r- ~i—i—|—i—i—r—i—|—i—i—i—i—|—i—i—r—I 280 560 *4 0Pi2 up+dn TDC] 1 1 1 1—520 600Fig. 37. Dot plots of pulse height in target- scintillator versus particle time of flight to the pion counters at (a) 60° and (b) 100°. These data were taken during the test run in September. The irp elastic scattering group is identified at the largest scattering angle. The continuum consists of events from 12C(ir,Trp) and 12C(ir,2p). A veto counter can easily eliminate the latter events.Experiment 324Polarization analysing power differences for in­elastic proton scattering from 12C at 400 MeV  (K.H. Hicks, TRIUMF; J.R. Shephard, Colorado)This experiment, to measure polarization- analysing power differences (P-A), has finished taking data and is completely ana­lysed. The results are presented in Table III. The ground state must have P=A by time-rever- sal invariance, and the collective states are expected to have P=A. These expectations agree with the data (within most errors) except at 14° which is being reanalysed. The 1+,T=0 and 1+ ,T=1 states at 12.7 and 15.1 MeV are calculated to have small, but measurable, P-A values. The 2+ ,T=l state at 16.1 MeV,32Fig. 38. Ratio of (ir+p) to (ir~p) differential cross sections from SWID [Arnst and Roper, VPI & State Univ. internal report CAPS-80-3, 1982 (unpublished)].which was the main emphasis of this experi­ment, is calculated to have a large P-A in both relativistic and nonrelativistic models. The data for these three states agree with these expectations. These data at a Tp (incident proton energy) of 400 MeV are the only measurements of P-A above T =150 MeV. The impulse approximation (IA) works best at Tp>400 MeV. The data are compared to IA calculations in Fig. 39. The agreement with calculations is fair, but gives hope that the nonlocality terms in the nuclear force for these models is handled fairly well. Complete theoretical interpretation of the data is in progress. The large P-A for the 16.1 MeV state is the only significantly nonzero P-A measured at an energy where the Impulse approximation is expected to do well.THEXAFig. 39. Measured P-A versus the proton angle for 12C(p,p') to unnatural and natural parity states. Calculations for nonrelati­vistic (DW81) and relativistic (DREX) impulse approximation models are shown.Table III. Polarization-asymmetry (error) for 12C(p,p') at 400 MeV.\ e *0 N. (°)\0.0(MeV)4.44(MeV)7.65(MeV)9.64(MeV)12.71(MeV)15.11(MeV)16.11(MeV)5.0 -.04(.04) -.04(.05) -.04(.13) +.10(.12) -.06(.06)8.0 +.04(.01) +.03(.01) -.04(.03) —.05(.04) +.09(.05) —.01(.04) + .44( .10)10.2 +.04(.01) +.05(.03) +.01(.02) +.05(.03) + .12(.05) -.09( .10) +.40(.06)11.8 +.02(.01) + .09(.03) +.04(.02) +.05(.02) +.22( .06) -.15(.10) +.37(.06)14.00 +.10(.02) + .06(.04) +.10(.03) +.06(.02) -.14(.15) -- +.54(.10)33Experiment 327Study o f the ( ir+,-ir*ir-) reaction on m0 , 2SSiand 40Ca at Tw = 240 and 280 MeV(N. Grion, INFN Trieste)The purpose of Expt. 327 is to study the energy and mass dependence of pion-induced pion production in order to obtain informa­tion on the contribution of the nuclear matter to pion production. At present there are several theoretical models which predict a total cross section ranging over more than two orders of magnitude, while there are no experimental data available on nuclei larger than deuterium (except for very few data from emulsion experiment).The experiment was very active during 1986. The main tasks that have been accomplished since the beginning of the year are the con­struction and testing of the second arm (CARUZ) for the coincidence experiment and the acquisition of data at 280 MeV incident pion energy on 160. A preliminary analysis of these data has also been completed.The CARUZ consists mainly of a stack of five plastic scintillators of different thickness. It has a solid angle of 0.25 sr. The energy of the pions stopping before the fifth scin­tillator is calculated on the basis of the light released in the stack itself. A test of the CARUZ was performed on the M13 channel in late February using four days of polarized beam. The channel was set at momenta ranging from 20 to 150 MeV/c, in order to acquire a calibration curve versus pion energy and to determine the efficiency for pion/electron discrimination. The average pion energyresolution between 10 and 55 MeV is betterthan 2 MeV. The discrimination of pions from electrons is achieved for more than 95% ofthe particles. During a later beam periodthe CARUZ was calibrated for protons using the proton component of the Mil channel when it is set for positive pions. The energy resolution was measured to be 3.2 MeV at 80 MeV proton energy. The CARUZ can mass identify pions from protons thus allowing the ( tt+ . tt+ tt* )  channel to be discriminated from the (it+ ,ir-p) one. The results of pion-elec- tron and pion-proton discrimination are shown in Fig. 40(a) and (b), respectively.During a two-week period in May-June the first data on pion-induced pion production reaction on 160 at 280 MeV was taken usingthe Mil channel. This experiment was com­pleted in July, and events at 240 MeV were also collected (data-taking will resume in April 1987). The outgoing pions were measured at opening angles between 70° and 180° and their energies measured in an interval that almost completely covers the allowed phase space. Negative pions were discriminated with the TRIUMF QQD spectrometer, while positive pions were mass identified with the CARUZ. Both pion trajectories were traced back to the target position by means of wire chambers.From a preliminary analysis of the 280 MeV experimental data a total cross section of1.5(0.3) mb for the (it ,2tt ) reaction mechanism was calculated. The error bar associated with the datum point accounts only for the statistics; systematic errors have not been included. Figure 41 shows the experimental value compared with several theoretical models (see references therein quoted) . The x points are theoretical calculations of Cohen and Eisenberg performed with the DWIA and taking into account possible nuclear spin-isospin mode enhancements by renormal­izing the one-nucleon transition operator. Their results are given for three different values of the Migdal parameter g1. The dashed curves are calculations of Oset and Vicente-Vacas, in which two- and three-point diagrams for the one-nucleon mechanism have been included. The main result of these calculations is that the enhancement of the cross section (from dashed curve a to b) is due to the binding of the outgoing pions in the nuclear medium, while there is no sizable precritical enhancement (from dashed curve b to c). Finally the continuous line is Rock- more' s theoretical predictions for the 180(tr+ ,ir+Tr+) reaction. The author calculates the (it,2tt) process in terms of the Fermi gas model of the nucleus and accounts for the effects of the incident pion attenuation on the one-body ttN * 2irN cross section. Rock- more's predictions are multiplied by a factor 7 in order to compare them with the experi­mental result which is for (ir+ , ir_ir+). The factor 7 arises from the ratio of the calcu­lated cross sections of Oset for the (ir+ ,Tr-7r+ and (ir+ ,TT+ir+) reactions.These preliminary results were presented at the APS 1986 Fall Meeting of the Division of Nuclear Physics, which was held in October in Vancouver.34T.O.F. (nstc)RUN W  55 MEVRUN 44 P  MEVTTilT: Tt* * “O  **• It* * It* ♦ ♦ OOCX : T V *  * “O  Tl~ + p  * 0 > *O )Fig. 40. (a) Pion and electron discrimination by the CARUZ. (b) Pion and proton discrimi­nation by the CARUZ.160 (TT + ,7T 7T + )T„ (MeV)Fig. 41. Integrated cross section for the reaction 160(tt+)tt~it+) at 280 MeV (□). Theo­retical calculation by Cohen et al. (x) [Nucl. Phys. A395, 389 (1983)]. Dashed and fullcurves are calculations by Oset et al. [Nucl. Phys. A454, 637 (1986)] and by Rockmore[Phys. Rev. C27, 2150 (1983)], respectively.35Experiment 3304~ stretched states in 160(O. Hausser, SFU/TRIUMF; R. Jeppesen, SFU;D. Frekers, Toronto)Preliminary measurements on inelastic scat— tering to the 4" states in ie0 at 200 and 290 MeV have recently been made at TRIUMF. These states are of particular interest because their nuclear structure is well known and because their form factors peak at large momentum transfer where the tensor and spin- orbit parts of the NN interaction are domi­nant. The stretched configuration of the 4 states (0p3/2 hole - 0d5/2 particle) implies that the transition has a single L transfer (L = J-l) and that the transverse and longi­tudinal form factors will be proportional. The transverse form factor, and consequently the longitudinal, can be determined from in­elastic electron scattering. The three 4 states in  ^8 0 (at 17.79, 18.98 and 19.80 MeV) are isospin mixed, but pion scattering mea­surements have been used to determine the isospin configurations. Only the effective NN interaction remains unknown and a measurement of the complete set of observables for these states allows one to get information on this interaction. Love and Klein [Proc. Sixth Int. Symp. Polar. Phenom. in Nucl. Phys., Osaka, J. Phys. Soc. Jpn. 55_, 78 (1986)] have used the PWIA to derive simple relationships between some of the spin observables and the effective NN interaction. These relation­ships, which do not have any nuclear struc­ture information in them other than the assumption of a stretched state, have been seen in the 200 MeV IUCF data and persist in full DWIA calculations.Preliminary measurements to demonstrate the feasibility of this experiment have recently been made at TRIUMF. Cross sections and analysing powers were measured at 200 MeV (20° < 9lab < 39°) and 290 MeV (16.5° < 9lab < 32.5°) with a resolution of 140 keV at200 MeV. The waterfall target developed by the University of Toronto group was used to keep the physical background below the 4“ peaks to a minimum. Cross sections were nor­malized by comparing the elastic peak in the water target data with the ^60 elastic peak in a mylar target. Preliminary analysis of the 200 MeV data indicates that cross sec­tions for the 19.80 MeV state agree with the previous data from IUCF [Olmer, Proc. LAMPF Workshop on Dirac Approaches to Nuclear Physics, LA—10438—C (1985)] (see Fig. 42).10? 10-1 •<0aM -2•a io10- 319.80 M eVo .  “  «o  o» •open — IUCF closed — E33015 25 35 45c o m  an g le  (deg)55Fig. 42. Measured cross sections for the 4“ state at 19.80 MeV of excitation in 160 com­pared to the data of Olmer. Error bars have been omitted on the TRIUMF data.A proposal to measure the complete set of spin observables using the MRS focal plane polarimeter and the longitudinal polarization on beam line 4B was submitted to the December EEC. These further measurements would be made at 350 MeV, the highest energy at TRIUMF for which all three components of the spin of the scattered proton can be reasonably determined.Experiment 335Energy dependence o f theJT-nucleus interaction from 1+ excitations in 2eSi(p,p')(O. Hausser, SFU/TRIUMF)We have studied inelastic proton scattering from 28Si at five incident energies (200, 250, 290, 360 and 400 MeV) to examine parts of the N-nucleus interaction which contribute to AS=1 spin-flip excitations. The 9.50 MeV state observed in the 28Si(p,p') spectrum of Fig. 43 is a rare example of a pure T=0 1+ excitation in nuclei. Its isospin purity can be inferred from the fact that it is not ex­cited in 28Si(e,e') [Schneider et al., Nucl. Phys. A323, 13 (1979)]. Together with thestrong T—1 state at 11.45 MeV it has been used as a probe state to study separately the36Fig. 43. Momentum spectrum of 250 MeV protons scattered from 28Si at 9qab = 4°.isoscalar and isovector components of the N- nucleus interaction. The experiments were carried out at the small-angle MRS facility using both polarized and unpolarized proton beams. The 200 MeV results were obtained in collaboration with C. Djalali, A. Galonsky and G. Crawley from Michigan State University. The cross sections at 200 MeV are larger by a factor of 2.03 than those from an earlier ex­periment at the same energy [Anantaraman et al., Phys. Rev. Lett. 52^ 1409 (1984)].From frequent inter-comparisons with the elementary pp cross section and from the reproducibility of individual runs we assess a systematic uncertainty of ±6% to the TRIUMF cross sections. This includes contributions from target nonuniformity and from errors in beam charge integration.®cu(deg)Theoretical cross sections were calculated in the framework of the DWIA with the following input parameters:1) Distortion effects were calculated with phenomenological optical potentials obtained from new elastic scattering data [Lin, MSc. thesis, Simon Fraser University, 1986] at 200, 250 MeV (see Fig. 44) and 400 MeV or from published work at 333 MeV [Hintz et al., Phys. Rev. C 3£, 1976 (1984)]. These poten­tials imply reaction cross sections in agree­ment with experiment.2) The widely used effective interaction of Franey and Love [Phys. Rev. C 31^ , 488 (1985)] was examined. Together with the distortion factors from 1) the dominant central part of this interaction implies ap n(0°,q=0)/B(GT) of 6.6 mb/sr at 200 MeV, 6.0 mb/sr at 250 MeV and 5.9 mb/sr at 400 MeV.3) Transition densities of Wildenthal and Chung [Chung, Ph.D. thesis, Michigan State University, 1976] were used which are in qualitative agreement with the distribution of isovector 1+ strength observed in (p,p') and (e,e') reactions. We note that the orbi­tal contribution to B(Ml, 11.45 MeV) is neg­ligible, thus a(p,p')/B(Ml) ~ (g|ff/g|ff)2. The more recent transition densities by Wildenthal and Brown [Wildenthal, in Prog. Part. Nucl. Phys. _11_ (Pergamon, Oxford, 1983) p. 5; Brown, private communication] overesti­mate the fragmentation of 1+ strength compared to both (e,e') [Schneider et al., op. cit.] and (p,p') experiments.®c«(deg)Fig. 44. Differential cross section (left) and analysing power (right) for 28Si(pT,p) elastic scattering at 250 MeV. The dashed lines represent calcu­lations using a nonrelativistic phenomenological optical model.371§\b(de«)£>ac•dSb■d10*10'10i(r1 1 I 1“ sifc.p ') 200 MeV10 ---------1 1 1 Ia Si(p,p’)  400 MeV9.50 MeV T=0 f(LF,WC)=0.46 in' 9.50 MeV T=0 f(LF.WC)=0.72.aa M —— — — -  ■ ■ ■ ■ ■ fc.............. •O  ,n-i \  10 .**’** S\ IS\\i i i ------- 1---------- 10-*\ ^ 1 1 1 t N» —e o 12 *«. (de«)IS s09(deg)12 15Fig. 45. Angular distributions for excitation of the strongest isovector (top) and isoscalar (bottom) 1+ states in 28Si(p,p') at 200 MeV (left) and 400 MeV (right).Angular distributions for the dominant iso­vector and isoscalar 1+ states at 200 MeV and 400 MeV are shown in Fig. 45. The theoretical curves are multiplied with 'quenching factors' f = cTgxp/otheory to obtain agreement with experiment. The 1+,T=0 distributions include the contribution of a weakly excited, unresolved 2+ state. The quenching factors are shown in Fig. 46 versus incident proton energy. The quenching factor for the 9.50 MeV state at 200 MeV is significantly smaller than those at 250 MeV and above. This may be an indication that density-dependent effects which have been neglected in our analysis areimportant below 250 MeV. The results for the11.45 MeV state are consistent with an average value f(p,p') = 0.78±0.0 at all ener­gies. The interaction of Franey and Love describes the energy dependence of isoscalar spin-flip excitations correctly.The high-resolution (e,e') results of Schneider et al. were used to calculate f(e,e') and feff(e,e') using free-nucleon g factors or effective g factor fitted to a large body of isovector magnetic moments and Ml transitions. These quenching factors are shown in Table IV together with double ratiosTable IV. Quenching of Gamow-Teller and isovector Ml strength for the 11.45 MeV state in 28Sif(p,p')a f(e,e')b feff(e>e') e ~ K M c\8sff/«8/ E'" ~ (i t )0.78±0.05 0.96±0.04 1.31±0.07 0.8110.07 0.6010.05^Average of the T=1 results in Fig. 46.Using B(Ml)eff = 4.07±0.22 and free-nucleon g-factors. cUsing renormalized isovector g-factors, (g|^/gg) = 0.859 and (g|^/ge) = 1.022 from Brown [Harrogate Conf. 1986].38R = f(p,p')/f(e,e') which remove most of the uncertainty resulting from the theoretical densities. From Rgff we deduce that (g^ )2is renormalized to (60+5)% of the free value, in agreement with (p,n) results in (s,d) shell nuclei [see Brown, Harrogate Conference 1986]. The renormalization of (gg /g5 ) is less severe, apparently because of additive contributions from meson exchange.Data for Fig. 46; WC transition densities.EP f(T=0) f(T-l)200 0.46±0.035 0.74±0.05250 0.70±0.042 0.84±0.05290 0.60+0.06 0.7010.06360 0.69±0.07 0.7710.06400 0.72+0.44 0.8410.05dashed line 0.69 0.78average (average of (average of4 points) 5 points)Experiment 337Measurement o f tensor observables in irdelastic scattering (G.R. Smith, TRIUMF)This experiment was designed to measure the tensor observables T20, T21 and T22 in the tt3 elastic scattering reaction, for incident energies spanning the (3,3) resonance. Theexperiment employs a scintillation counter array, depicted schematically in Fig. 47, which allows measurements at 6 scattering angles to proceed simultaneously. One novelty of this (single scattering) experiment was the first use of a tensor polarized deuteron target in a hadronic scattering reaction.In last year's annual report we reported that initial measurements of T2Q had just been completed at T^ = 134 and 151 MeV. In themeantime these results have been published [Smith et al., Phys. Rev. Lett. 57^ 803(1986)]. The objective of this first run was to explore the angular dependence of the ten­sor analysing power T2Q in the energy region where strong angular oscillations of thetensor polarization t20 were observed at SIN.The SIN results have been cited as evidence150  200  250  300  350  400  450 Ep (MeV)Ep (MeV)Fig. 46. Quenching factors for the strongest isovector (left) and isoscalar (right) 1+ states in 28Si between 200 and 400 MeV. The dashed lines are average values.for dibaryon resonances. As Fig. 48 shows, our (single scattering, tensor polarized tar­get) T20 results are in agreement with (double scattering, unpolarized target) mea­surements of t2Q made at LAMPF and at TRIUMF, but are inconsistent with the SIN (double scattering, unpolarized target) results.A parallel investigation [Smith et al, Nucl. Instrum. Methods (in press)] performed during this experiment involved a direct measurement of the magnitude of tensor polarization in the polarized deuteron target. This calibra­tion measurement was made by utilizing the known tensor analysing power at 90° c.m. in the to 2p reaction. The measurement was made for a variety of experimental conditions at an incident bombarding energy of 80 MeV. The results were consistent with those de­duced from NMR measurements of the target vector polarization. An investigation of the rf-burning technique for producing enhanced deuteron tensor polarizations revealed that this technique was ineffective for producing enhanced tensor polarizations with reasonable relaxation times.39Mil ir* BEAMFig. 47. The experimental layout is shown, with the pion beam incident from the top. The pion detector rings are used to define the solid angle and record the TOF and energy loss of scattered pions. The deuteron detec­tor rings record the energy loss, TOF and total energy of recoil deuterons, as well as vetoing minimum ionizing particles.First measurements of the tensor analysing power T21 were obtained in the spring of 1986. The resulting 12-point angular distri­bution at T-jy = 180 MeV has been submitted for publication. These data are in remarkably good agreement with Faddeev calculations from all the theoretical groups currently working on this problem. The situation is in marked contrast to that for T20 where the data can only be described when effects originating from pion absorption are either dampened or left out entirely.Measurements of T2q at 180, 220 and 256 MeV were also a product of the spring 1986 run. These data have been analysed. The agreement with the few existing tensor polarization data from LAMPF is impressive. The effects of pion absorption are predicted to be more pronounced at these higher energies, and the data bear this out. Agreement with the data is only achieved by Faddeev calculations inFig. 48. The tensor analysing power angular distributions (solid squares) are compared to the existing tensor polarization data from SIN (open triangles), from LAMPF (open circles) and from TRIUMF (open squares) at (a) 134 MeV and (b) 151 MeV. The conversion of T20 (c.m.) to t2Q (lab) is down by admix­ing calculated T21 (c.m.) and T22 (c.m.)values to the measured T2g.which pion absorption is left out. The full calculations are smaller in magnitude than the experimental data.The experimental program to measure the ten­sor observables in ird elastic scattering will be completed with a run taking place in December. The objective of this run is to measure the tensor analysing power T21 at 134 and 220 MeV. For this run a Jll Starburst has been implemented for data acquisition. The Jll basically multiplexes the experiment in such a way that the PDP 11-34 reads from the Jll memory only those words pertaining to the detector arm which has generated the event. This scheme results in a factor of six fewer magnetic tapes required for the experi­ment, with a corresponding increase in the speed with which the data can be analysed off line.Measurements of T21 and T2Q at 256 and 294 MeV, as well as measurements of T22 be­tween 134 and 294 MeV, were performed at SIN during the course of the year. The new results from this experiment at TRIUMF, together with those from SIN, form a complete40set of tensor observables in the tt3 elastic scattering reaction for energies spanning the (3,3) resonance.Experiment 344Excitation o f stretched particle-hole states incharge exchange reactions (J. Watson, Kent State)In the summer of 1986 data were taken for the 28Si(p,n)28P(4.95 MeV, 6") and 88Sr(p,n)88Y (1.48 MeV, 9+ ) reactions at 300 and 400 MeV. The states studied are 'stretched' high-spin particle-hole states which are excited prim­arily by the isovector-tensor part of the nucleon-nucleon interaction. A stretched state has both the particle and hole in stretched orbits (jp = £p+1/2; = f.^ +1/2)with their angular momenta coupled to the largest possible value (J = jp + + 1). Astretched state typically has a very pure configuration because there are no states within 2 ■tfw of excitation that can mix with it due to its high spin.Data were taken at TRIUMF at 300 and 400 MeV in the angular range from 18° to 40° which covers momentum transfers from about 1.0 to2.5 fm-1, where the isovector-tensor force is the dominant isovector part of the nucleon- nucleon interaction. This required a special wide-angle configuration of the CHARGEX (p,n) facility on the MRS so that measurements could be made at angles larger than 30°.Measurements on these same stretched (p,n) transitions are under way at the Indiana Uni­versity Cyclotron Facility at 100 and 200 MeV so that the energy dependence of the excita­tion of these states can be studied over a large energy range.Replay of the data tapes is proceeding; an off-line neutron-energy resolution of about1.0 MeV has been obtained. Figure 49 shows off-line spectra at 28°. Measurements of stretched particle-hole states excited in (n,p) reactions are scheduled for the winter of 1987.Experiments 350/351Study o f the energy dependence and the A dependence of the double charge exchange reaction at low  energy (A. Altman, Tel-Aviv)The DCX (double charge exchange) reaction via the DIAT (double isobaric analog transition) at low energies (~50 MeV) has recently re­ceived a great deal of interest because itsamples the multi-nucleon aspects of the pion nucleus interaction where the SCX (single charge exchange) reaction on the nucleon is very small. At TRIUMF the 180(tt+ , it- ) 18Ne re­action was studied previously [Altman et al., Phys. Rev. Lett. JI5, 1273 (1985)]. Several theoretical models [ LAMPF Workshop on Pion Double Charge Exchange, LA-10550-C (1985)]have been developed that more or less explain this data and the 1LtC data.In order to test the ability of these models to reproduce the A dependence of the reaction at 50 MeV, data have now been taken on larger nuclei (26Mg and 56Fe). Figure 50 shows the angular dependence of the 50 MeV DCX reaction on 26Mg. Only a single angle (30°) was studied for 50 MeV on 56Fe. Figure 51, where *4C and 180 data have been included, shows the resulting A dependence of the extrapo­lated zero degree cross sections for DCX at 50 MeV.4030ini—§ 20ooJ O030ini—i 20oo1000 100 200 300 400FOC AL P L A N E  POSITIONFig. 49. Neutron spectra at 28° from the 28Si(p,n)28P and 88Sr(p,n)88Y reactions at 300 MeV. At this angle the 28P(4.95 MeV,6“) and 88Y(1.48 MeV, 9+) stretched particle-hole states are excited strongly.410c.m . ( d e g )Fig. 50. Angular dependence of the DIAT for 50 MeV pions on 28Mg.180 (7 T + ,7 T - )18N e  , D I A TFig. 52. Energy dependence of the DIAT on 180. The solid curve is a polynomial fit to the 49 MeV data explained in Altman et al. [Phys. Rev. Lett. _55^  1273 (1985)].The data have been multiplied by the factors shown on the right of the figure.Fig. 51. A dependence for the DIAT for 50 MeV pions. The solid line shows the p j-10/3 dependence that is seen at higher energies.l80 (7 T + , 7 r " ) 18N e  , D I A TFig. 53. Energy dependence of the data of Fig. 52 extrapolated 0° and at higher ener­gies (>80 MeV) the 5° cross sections from Seidl et al. [Phys. Lett. 154B, 255 (1985)].42The theoretical models mentioned above must also be able to explain the energy dependence of the reaction in the region of 50 MeV, so data have been taken for the DCX reaction on 180 for energies from 24 to 80 MeV. The results at 80 MeV confirmed those taken prev­iously at EPICS [Greene et al., Phys. Rev. C 25, 927 (1982)]. Figure 52 shows the angular distributions taken at each of the energies studied. It is noteworthy that these distri­butions appear to be becoming steeper as the energy is increased. Figure 53 shows the en­ergy dependence of the DCX reaction on i80 as extrapolated to 0° from the angular distribu­tions of Fig. 52. The crosses and stars in this figure are data taken at 5° with EPICS at LAMPF [Seidl et al., Phys. Lett. 154B, 255 (1985)].Experiment 354Study o f nuclear structure and density dependenceof the effective interaction for N =  50 isotones(D. Frekers, Toronto)An attempt is being made to extract neutron transition densities for a number of low- lying levels in the N=50 isotones 88Sr and 89Y using intermediate elastic and inelastic proton scattering. Both nuclei provide an ideal test case because several of the neu­tron transition densities are expected to be substantially different from their proton counterparts both in shape as well as in strength. Furthermore, proton transition densities to various levels are accurately known from electron scattering experiments [Schwentker et al., Phys. Rev. Lett. fiC^, 15 (1983)] and can serve as input parameters in distorted wave calculations. Previous calcu­lations have also indicated sufficient sensi­tivity to the shape of the neutron transition densities.We have focused our studies on the two low- lying 2+ states at 1.836 and 3.219 MeV and the 3“ state at 2.734 MeV in 88Sr, and the 9/2+ , 3/2”, 5/2 and 5/2+ states at 0.909,1.507, 1.745 and 2.222 MeV in 88Y. For the 2+ states in 88Sr electron scattering experi­ments indicate a mixture of it(2pj/2 >If 5 /2*1) and tt(2p1/2 ,2p3/2_1) single particle wave functions. Core polarization effects from the valence neutrons are, however, unknown and are subject to our study. The 3“ state is highly collective and here proton and neutron transition densities can be assumed to be largely identical. 88Y can be regarded to a large extent as a 88Sr core with an additional unpaired 2p1/2 valence protonwhich, for the 0.909 MeV state, gets promoted into the lg9/2 orbit. In fact, the 5/2+ state at 2.222 MeV can be understood in the weak coupling limit as an octupole core vibration coupled to the 2p1/2 proton. However, for the 3/2- and 5/2“ states at 1.507 and 1.745 MeV electron scattering data do not support the picture of the coupling to either of the two low-lying 2+ states in 88Sr. The measured proton transition densities have nothing in common with each other or either of the 2+  states in 88Sr. This apparent failure of the weak coupling model is at present not under­stood, and proton scattering may well be used as an additional means to elucidate the situation.We have measured cross section and analysing power angular distributions for both nuclei at 400 MeV, complementing previously taken data for 80Zr at the same energy. The raw data are shown in Figs. 54 and 55 for 88Sr and 89Y, respectively. It is seen that the angular distributions for the two 2+ states in 88Sr are radically different, and as ex­pected the data for the 3~ state in 88Sr and the 5/2+ state in 88Y are largely identical. However, even the data for the first 2+ state in 88Sr are apart from a scaling factor in cross section similar to those of the 3/2“ and 5/2” states in 88Y. In view of the large differences seen in electron scattering this result is unexpected and seems to point to large neutron contribution from core polari­zation effects.Experiment 357Measurement o f B h  f o r 19F and 26M g using the(n,p) reaction (W.R Alford, Western Ontario)Cross sections for *8F(n,p) and 28Mg(n,p) were measured at angles of 0°, 5° and 10° at 198 MeV. For 28Mg we find L=0 strength con­sistent with shell model predictions, assum­ing the same quenching of the free-nucleon GT operator as in the (p,n) reaction. In addi­tion, considerable L=1 strength is found in the region 4 to 12 MeV excitation.For 19F a shell model calculation predicts little GT strength. Our results are general­ly consistent with the calculations but analysis of the results has been complicated by the appearance of strong transitions with L=1 at low excitation and by impurity groups from the carbon in the Teflon targets used in the measurements.43-QJH,a10  103 102 1 0 1 1 0 °  i o _1 io—210"*  1 .0  0.6 0.2 —  0.2 —  0.6 —  1.0,I  8 8  S r ( p , p )  ® ® S r ( g . s . ) .  t  E p =  4 0 0  M e Vr^ \  ~20T S ^ V T S\ : V  \  •jj 1_! Li_1 0■ 10' I 10°§Kf'b _2*101010‘ to' Ay 0.6 0.2 -01  -0.6 -to.'N(1):/\‘ V\AM\A(2)sY \,0 10 20 30 40 50 *0_»r m10 20 30 40 50§cjn.[deg.],10to'0.60.2-02- 0.610,0 102 0  3 0  4 0©  c . m .  [ d e g . ]10’10°10-10"!10_i10^10"“-I20 50 40 50 , „0 —m r 1- toT H T T aI 010.2 - 0.2 - 0.6 -to,(3)10 20 30 40 5020 30 40 50 8 cm- [deg.]tO 20 30 40 508 cm [deg.]Fig. 54. Cross section and analysing power angular distributions for elastic and inelas­tic proton scattering from 88Sr at 400 MeV. The excited states are (1) 1.84 MeV (2"f); (2) 2.74 MeV (3^); (3) 3.22 MeV (2+).B10-1031 0 21 0 110°  —1\ ®9 Y (p,p)89Y (g.s.)_ :  _  E 0  =  4 0 0  M e V\ r \V/~I O  2 0  3 0  4 0J V \ r \  \ r0 » 20 30 40 50*cmld»]0 0 20 30 40 50 D 20 30 40 50 8cm[deg]1,1 /  V \:  s -V ,___1___ 1___ 1___ 1___Ulr— t —w -ti 1 1to*— 1— u  0 t t  201 1 :30 40 50Fig. 55. Cross section and analysing power angular distributions for elastic and inelas­tic proton scattering from 89Y at 400 MeV.The excited states are (1) 0.91 MeV (9/2+);(2) 1.51 MeV (3/2*); (3) 1.75 MeV (5/2*);(4) 2.22 MeV (5/2+).Experiment 359The spin response for 54Fe (p,p') at 290 MeV(C. Glashausser, Rutgers; O. Hausser, SFU/TRIUMF)This experiment examines the distributions of spin-flip strengths in selected f-p shell nuclei (^Ca and 54He) up to excitation ener­gies of 45 MeV using the (p,p') reaction at 290 MeV. We report here measurements of the spin-flip probability Snn and analysing power Ay for 54Fe(p,p') at five angles between 3.1° and 15°. The experiment was carried out with the focal-plane polarimeter of the MRS at a single B-field setting. In Fig. 56 a spec­trum of the spin-flip cross section aSnn is compared to a cross-section spectrum. An accumulation of Gamow-Teller strength (AS=1, AL=0) is clearly observed near Eexc = 10 MeV. The background of AS=0 natural parity states seen in the a spectra is absent in the oSnn spectra.At 9 = 3.1° the integrated spin-flip cross section between 7.4 and 12.6 MeV is 3.5 mb/srof which 80±10% is estimated to be AL=0. A 'quenching factor'f = J aSnn(exp)/J aSnn(theory) = 0.39±0.06has been deduced. The reaction calculation was carried out in the framework of the DWIA, using the Love-Franey interaction and assum­ing a simple 5‘tFe(f7/2)llt' ground state. The distorting potentials in this calculation are based on detailed measurements of a and Ay for 5‘*Fe(p,p) at 290 MeV. These elastic data are shown in Fig. 57 together with predic­tions from the density-dependent Hamburg interaction [Dymarz, private communication]. The optical potentials have been reduced empirically by 20% to improve the agreement with the data.The quenching factors from the (p,p'), the (p,n) [Rapaport et al., Nucl. Phys. A410, 371 (1983)], and (n,p) [Expts. 267, 283 Vetterli and HSusser] reactions are shown in Fig. 58 The 51V(p,p') datum was deduced from our cross-section data at 3°, with corrections44-Q  (MeV)- 5  0 5 10 15 20 25 30 35 40 45-Q  (MeV)Fig. 56. Doubly differential cross section (top) and spin-flip cross section (bottom) for the reaction 51tFe(p,p’) at 0 = 3.1° and Ep = 290 MeV.for the AS=0 background derived from the oSnn and a data for 51*Fe. The large quenching observed may be explained by the severe trun­cation of the shell model basis used. The results also show that of the three reactions the (n,p) reaction is most severely affected by ground-state correlation.Fig. 58. Quenching factors for (p,n), (p,p') and (n,p) reactions on (f,p) shell  angleFig. 57. Differential cross section (top) and analysing power (bottom) for proton elas­tic scattering from 54Fe at 290 MeV. The theoretical curves are discussed in the text.The data at the larger angles are being used in a multipole decomposition of the angular distributions and in a search for Gamow- Teller strength at higher excitation ener­gies. They are also of particular interest when compared to the surface RPA model of Smith and Esbensen [private communication]. The spectra for Snn and Ay at 15° are shown in Fig. 59 together with the quasielastic RPA response (solid line) and the free NN response (dashed line). We note that the residual interaction in the surface response theory introduces a distinct slope to Snn which is in excellent agreement with the data. Such close agreement might be expected since Snn is a remarkably sturdy quantity which is unaltered by such factors as distor­tions, two-step processes, effects of relati­vistic dynamics and of Fermi motion averag­ing. The surface response prediction for Ay451.00.80.6,0 .40.20.0-.254 Fe(p,p) 290 MeV \  15.0 deg.....1.0 MeV b in sI 1 i 1 u- 5  0 5 10 15 20 25 30 35 40 45%xc (MeV>Fe(p,p) 290 MeV 15.0 deg.1.0 MeV b in s' - 5  0 5 10 15 20 25 30 35 40 45EeXc (MeV)Fig. 59. Analysing powers (top) and spin- flip probabilities (bottom) observed at 9 = 15°. The solid line represents the surface RPA response theory of Smith and Esbensen, the dashed line was obtained from the free NN response.exhibits again a strong slope versus excita­tion energy but consistently overpredicts the data by 30-40% at all angles. We emphasize that a reduction in Ay of the required amount is predicted by Dirac calculations [Horowitz and Iqbal, Phys. Rev. C 13, 2059 (1986) and private communication] of quasielastic scat­tering, where it is a consequence of the re­duction of the effective nucleon mass in the strong scalar field of the nuclear medium. A search for additional manifestations of this m*-effeet in other spin observables is the subject of future experiments (Expts. 397, 431).Experiment 365 A search for the tetraneutron (T. Gorringe, UBC)Experiment 365 is searching for the produc­tion of tetraneutrons (nuclei containing four neutrons but no protons) in the pion double charge exchange (DCX) reaction ^He^-, ir+)4n. Production can be recognized by a monoener- getic peak in the outgoing tt+ energy spec­trum. To measure this spectrum we are using a 170 MeV/c it- beam from M9, a high pressure helium gas target and the TRIUMF time projec­tion chamber (TPC). During April ir+ energy spectra were measured, for a total of 2.0x1011 7r-'s, incident on both full andempty target flasks. Equal statistics for the target-empty measurement were obtained because of the large it4" background from DCX on the aluminum walls of the target flask. In addition several ir+ elastic scattering runs were taken for calibration.Analysis of these data is currently under way. Figure 60 shows the present ir+ momentum spectrum from the helium target flask after cuts have been imposed to eliminate the in­tense backgrounds from protons and positrons. These cuts are based on particle energy loss within the TPC, and through the assorted trigger counters. The protons are easily removed; however, a small positron contamina­tion still remains. In Fig. 61 is shown the same ir+ momentum spectrum after subtraction of the empty target background. The overlaid peak shows the expected signal if tetraneu­trons were produced with a cross section of 50 nb/sr, with zero binding energy. The width of the peak is determined by the tt~ beam momentum byte. The counts below the peak are due to continuum pion double charge exchange on the helium. As to the few counts in the region that would signal tetraneutron forma­tion, they may be pions, or possibly positron contamination, since in this region if/e iden­tification is particularly difficult. We are presently trying to answer this question.If we can conclude these events are pions, and that positron contamination can be elim­inated, then we would plan to take further data to take a closer look at this energy region. If no, and positron contamination limits our sensitivity, then a limit of about 10 nb/sr can probably be set upon the tetra­neutron production cross section.46GGoo4 He(iT ,7T+ )4nexiting i&  incident nm o m e n t u m  m o m e n t u m\f BE(4n)=0 MeV 170 MeV/c%Inan40 60 80 1007r+  M o m e n t u m  ( M e V / c )120 140 160 180 200Fig. 60. ir+ momentum spectrum before target- empty background subtraction.4 He(mr ,7i+ )4 n86-4 -2 0 —2 Hpeak corresponding to ^n production cross section of 50 nb/sr—6 -|------- 1------- r40 60 80 100 120 140 160 180 2007T+  M o m e n t u m  ( M e V / c )Fig. 61. Tt+ momentum spectrum after target-o m n f u  K a r V o r n n n H  cnkfrflrrion.Experiment 366Measurements o f d a /d ila n d  Am to exclusive states of 160 (p ,iry70 between 250 and 500 MeV (P. Walden, TRIUMF)Experiment 366 received eleven 12 h shifts of polarized beam from September 18-24 on the MRS spectrometer. Angular distributions of da/dfl and analysing powers (Apjg) were taken at 354 and 250 MeV. A partial angular dis­tribution was also taken at 489 MeV. Prior to the Expt. 366 run a 160(p ,tt+)170 run at 200 MeV was run on the QQSP spectrometer at IUCF in a collaborative effort.signment, it being part of aconfiguration of which the 11/2 state is a member. The peaks at ~15.95 MeV and ~17.4 MeV have not yet been identified with any known state. There is speculation that these states could be the 11/2“ and 13/2” members of a ^p3/2^_1 (d5/2^2 The relation­ship of these highly excited states with similar excited states [Korkmaz et al., IUCF preprint, submitted to Phys. Rev. Lett.] seen in (p,u+) to 13C and 14C remains for further analysis of the data. The peaks above the17.4 state are states of 8Li, present because a LiOH target was used.Thus for the first time angular distributions of both do/dfi and ANq have been obtained for a (p,tv) reaction at several energies in the region of 200-500 MeV. Observing the energy dependence of these parameters will it is hoped be a key to understanding the reaction mechanism as well as a key test of the theo­ries. The energy region is ideal, as the role of the A should be dominant in this region.A sample 170 excitation spectrum is shown in Fig. 62. The three prominent peaks at lower excitations are identified with the 5/2+ ground state, the 5.218 MeV (9/2“) state and the 7.757 MeV (11/2“) state. The energy reso­lution of the spectrum is between 200-300 keV The assignment of 9/2" for the 5.218 state is not yet firm, but its strength in this reac-. tion [Ziegler, Ph.D. thesis, Univ. of British Columbia (1985)*] would indicate such an as- Evidence is shown here that (p,ir+) cross sections have an overall global weighting of 2 J+l.EXCITATION ENERGY(MeV)Fig. 62. An 170 excitation spectrum from the 160(p,ii+)170 reaction.471.00.5oz 0 <-0.5□  157 MeV 16, O 200  MeV A 2 50M eV  ~  *3 5 4 M eV  A 4 89M eV500 700qc m .(MeV/c)—  1q c.m .(MeV/c)Fig. 63. Analysing power as a function of momentum transfer for the 160(p,n+)170 reac­tion to the 5/2+ ground state of 170. Lines are drawn to connect points of similar ener­gies. The 157 MeV data is from Sjoreen et al.In Fig. 63 the analysing power Ajjq f°r 160(p,ir+) 170(g. s . ) at several different ener­gies is presented as a function of momentum transfer. Included in this plot are Aj^ q from the only previous experiment [Sjoreen et al., Phys. Rev. C 2A_, 1135 (1981)] to measure this parameter as well as preliminary results from the recent IUCF run at 200 MeV. It is inter­esting that the analysing power is somewhat constant in shape and oscillatory in behavi­our. The theoretical predictions for this Afgo (shown in Fig. 64) are from Cooper and Matsuyama [TRIUMF preprint TRI-PP-86-25, sub­mitted to Nucl. Phys. A] who have employed a relativistic stripping mechanism and a A-hole model. The theory shows qualitative agree­ment with the data. The signs, the number of oscillations and the magnitude of the oscil­lations are consistent with experiment. It is hoped improvements can be made to the model to achieve better agreement. However, this should not detract from the remarkable agreement with the energy behaviour of the data such as it is.Experiment 373Low energy pion scattering and pionic atom  anomaly (D.R. Gill, TRIUMF)Masutani and Seki showed recently [Phys. Lett. 156B, 11 (1985)] that the anomalousshifts and widths of atomic levels seen in some pionic atom studies could be fit with a set of optical model parameters that were not compatible with the parameters required to fit the majority of such data. They alsoFig. 64. The energy dependence of the anal­ysing power as a function of momentum trans­fer for the reaction 160(p,tt+ ) 170(g.s .) from the theory of Cooper and Matsuyama. The line designations are: dotted 200 MeV, short-dashed 250 MeV, long-dashed 350 MeV, solid 450 MeV.demonstrated that the existing low energy elastic scattering data were not sufficient to rule out such a modification of the opti­cal model parameters. A study of the elastic scattering of 20 MeV negative pions from ^Ca was therefore undertaken in an attempt to resolve this discrepancy.Experiment 373 received 11 shifts of beam in August. Data were taken at 20 MeV for posi­tive and negative pions elastically scattered from 12C and lf0Ca over the angular range from 45° to 125°. The analysis of the data is not yet complete, but cross sections should be available early in 1987.Experiment 374Non-analog DCX, the K0 ( ir+, r ) 16Ne reaction(D.R. Gill, TRIUMF)G. Miller showed that a calculation in which the valence neutrons in nuclei such as 14C and 180 were assumed to exist for some frac­tion of time in a six-quark bag state could explain the double isobaric analog transition cross section seen in 50 MeV pion double charge exchange (DCX). Since then several other theories have arrived on the scene that can equally well explain the data. Most of these theories achieve the required cross sections by including in the calculations transitions through excited states of the core of the nucleus in question. An experi­mental study of the DCX reaction on the core48of such a nucleus is thus indicated. Such a study has been undertaken with 50 MeV pions in 160 (the core of 180).Experiment 374 received beam in late December with the remainder of the data-taking to occur in January 1987. Cross-section results will become available early in 1987.Experiment 376Study o f the 90Zr(n,p) reaction a t 200 MeV(S. Yen, TRIUMF; M. Moinester, TRIUMF/Tel-Aviv;B. Spicer, Melbourne)This experiment is the first to study the (n,p) reaction at intermediate energies on a heavy nucleus where the normally dominant Gamow-Teller resonance is, to first order, Pauli blocked by the neutron excess. Whatever GT strength does exist would be a sensitive measure of ground-state correlations in 90Zr. A significant GT strength in the (n,p) chan­nel would invalidate the tentative conclusion based on RPA calculations [Osterfeld et al., Phys. Rev. C 3^, 372 (1985); Klein et al., Phys. Rev. C 31_, 710 (1985)] that no A-hole admixtures are required to explain the appar­ent quenching in the 90Zr(p,n) reaction. The Pauli blocking of the GT also offers a 'win­dow' through which we may observe spin iso­vector giant resonances. The (n,p) reaction is favourable for giant resonance studies because the lower Q value for the reaction, as compared to (p,n), means that the states-  Qnp MeVFig. 65. 90Zr(n,p) spectrum at 200 MeV.will be narrower, more concentrated, and thus easier to observe.During May data-taking for Expt. 376 was com­pleted during 21 shifts of beam time. Data were taken for 200 MeV incident neutron ener­gy at lab scattering angles of 0°, 3°, 6°, 9°, 13°, 17° and 21°. A sample spectrum at 0°, with instrumental background subtracted, is shown in Fig. 65. On the basis of the angular distribution there appears to be little Gamow-Teller strength. A major sur­prise is that the spin giant dipole resonance is far less prominent than predicted [Klein et al., op. cit.]. Off-line analysis of the data is in progress, with the aim of extract­ing the various giant multipole resonances which make up the spectrum.Experiment 377Measurement o f the charge symmetry parameterA-n-in the n d  elastic scattering reaction(G.R. Smith, TRIUMF)The aim of this experiment was to provide an accurate measurement of the asymmetry param­eter Aw in the ird elastic scattering reac­tion over the energy region spanned by the (3,3) resonance. The asymmetry parameter A^ is defined as the ratio A^ = (a~-a+)/(a“+a+), where the differential cross sections a+ and o~ refer to the ir+d and Tr~d elastic scatter­ing reactions, respectively. Previous mea­surements of this quantity have been per­formed at LAMPF, at bombarding energies of 143 and 256 MeV, using a pion spectrometer. The LAMPF results at 143 MeV indicated the existence of a (marginal) enhancement in A^ near 100° c.m. at 143 MeV. Calculations reproducing this enhancement were presented at the Versailles PANIC conference. There­fore, one goal of Expt. 377 was to study this region in an attempt to determine whether this enhancement really exists. The second goal of the experiment was to provide mass and width differences of the A isobars, which are responsible for predicted differences in TT+ d and ir-d elastic scattering cross sections based on Faddeev calculations. The energy dependence of A^ is an essential ingredient in this aspect of the experiment.The data collection for this experiment has been completed during the course of two run­ning periods in 1986. Twelve-point angular distributions of A^ were obtained at pion bombarding energies of 140, 180, 220 and256 MeV. Roughly half of the beam time was used to explore the consequences of possible49systematic errors on the outcome of the experiment. Numerous consistency checks were made, but the two dominant experimental dif­ficulties associated with measurements of this kind were studied in great detail.The first difficulty is associated with the different incident beam rates for each pion polarity. The ratio of incident beam flux for tt+ and ir” varies between roughly 7 and 30 on the Mil channel, depending on the bombard­ing energy. Four options are available to the experimentalists, if one varies one, both or neither of the two channel variables asso­ciated with the pion flux at a time. These two variables are the channel vertical (in­tensity) slits and the pion production tar­get. Cross sections and were measured under each of these four conditions for selected kinematical conditions in order to determine the effects of possible systematic uncertainties associated with this effect.The second difficulty is associated with the different fraction of pions in the beam for tt+ and ir”. This correction, arising mainly from tt° decay in the vicinity of the pion production target, gives rise to more elec­trons being present in the ir” beam than positrons in the ir+ beam. Using cyclotron rf referenced TOF down the length of the Mil channel, it is easy to determine this ratio for the energies studied in this experiment. However, the 2 ns wide beam bucket at TRIUMF makes it impossible to cleanly separate the muons from the pions at the higher energies. Therefore, considerable attention was paid to confirming experimentally the hypothesis that the p+/ir+ ratio was the same as the p”/ir” ratio. These ratios were measured at several lower bombarding energies where the muons can be separated easily from the pions in the TOF spectra. The results of this test demon- trated that the difference in muon contamina­tion for tt+  and ir- is less than 1%.The technique employed for the measure­ments was similar to that used for Expt. 337 (see p. 39). Briefly, a six-arm scintillation counter array was used to detect the scat­tered pions and associated recoil deuterons in coincidence. This technique has proven to be a highly precise method of determining relative cross sections for both ird and irp elastic scattering reactions. Some typical two-dimensional spectra obtained at 180 MeV are shown in Fig. 66. The ird yield is deter­mined by counting the number of events fall­ing within the polygon in this figure. The data analysis for the experiment should be completed early in 1987.Experiment 379Energy dependence of the (p,n) cross section for 13C and 15N (W.R Alford, Western Ontario)Angular distributions from 0° to 20° have been measured for the (p,n) reaction on 7Li, 12C and 1 at 200, 300 and 400 MeV. This is the first attempt to obtain a reliable cross- section measurement for the (p,n) reaction, and considerable effort was directed to checking the reproducibility of measurements. With an achromatic beam tune it was found that measurements could be routinely repro­duced to within 2 or 3%. The analysis of the measurements is in progress at this time.Experiment 381Measurement o f spin observables using the (p,p"Y)reaction(K.H. Kicks, TRIUMF; J.R. Shepard, Colorado)This experiment looks at the (p,p'y) reaction for inelastic proton scattering. Very encour­aging results were obtained from a (p ,p 'y ) test run on January 18-22. The set-up for this run is shown schematically in Fig. 67. The y detector was placed at 78° and about 9 in. from the target and shielded around with 2-3 in. of lead inside 1-3 in. of plas­tic (to help stop neutrons). The beam dump was an iron block located about 1 m from both target and detector. Data were obtained at 200 and 400 MeV at coincldent-proton angles of 5° to 9° using the MRS along with a BG0 detector.A single spectrum for 12C(p,py) at ©mrs = at 200 MeV is presented in Fig. 68(a), where the elastic peak has been prescaled by a fac­tor of about 250. The 1+ peak (at 15.1 MeV) is quite prominent. This singles spectrum may be compared with a 1 2 C ( p , p ' y )  coincidence spectrum (0 = 5°, 200 MeV) in Fig. 68(b).Only the 1+ state at 15.1 MeV remains since the other states (and the continuum states) do not decay strongly through y emission. Figure 68(b) is gated by the 'reals' peak in the time-of-flight (TOF) spectra started by the MRS trigger and stopped by the (delayed) Y detector time signal. Figure 68(c) shows the proton spectra gated by the 'randoms' peak in the TOF spectra; the TOF spectrum is shown in Fig. 68(d). The elastic peak (which cannot y decay) has about the same magnitude in both reals and randoms coincident spectra. We estimate these random coincidence are only 10% of the coincident data judged by the ratio of the 1+ peak to the elastic peak for both singles and coincidence proton spectra for a beam current of about 1 nA.50I " " I " 1 T rTT 200 440 680 920 1160 1400 D1+D2 ADC200 440 680 920 1160 1400 D1+D2 ADC1100 ■t, 1000 -o^  9005Jj 800 -1_L i_L _L+++,700600f + + + V + * / + tVr O  +7T+ C'I 1 1 1 1 I I200 440 680 920 1160 1400 D1+D2 ADC1100 -1000 -o -H900 -EQ :1(N 800 -0, -700 -600 -7T C11111111111111111 rr r200 440 680 920 1160 1400 D1+D2 ADCFig. 66. Typical two-dimensional spectra at 180 MeV. The vertical axis corresponds to the TOF difference between scat­tered pions and recoil deuterons. The horizontal axis shows the sum of the pulse heights in the deuteron AE and E coun­ters. The upper two scatterplots were obtained using a CD2 target and the lower two scatterplots using a graphite back­ground target. The foreground and background spectra here have been normalized to correspond to the same number of car­bon nuclei and incident pions. The polygon encloses the area corresponding to wd elastic scattering. The other bands cor­respond to protons detected in the deuteron arm.We conclude that the coincidence data show a clear signal for y decay of the 15.1 MeV 1+ state in 1ZC at incident proton energies of 200 and 400 MeV. The random coincidence arefs .n  eicctFig. 67. Set-up for the (p,p'y) test run where both BGO and BaF2 detectors were exam­ined .small and easily subtracted from the reals spectra, even under the very adverse condi­tions for this test with the beam dump only 1 m away and with minimal shielding.The set-up for the full (p,p'y) experiment is shown in Fig. 69. The design and fabrication of all parts for this set-up is completed. The set-up included 8 in-plane and 2 out-of­plane y detectors. This set-up will be tested in January 1987 using 3 existing BGO detectors from our Kentucky and Los Alamos colleagues.Experiment 383A test o f Gamow-Teller sum rule using chargeexchange reactions on 54Fe(M. Vetterli, O. Hausser, SFU/TRIUMF)Charge exchange reactions at intermediate en­ergies have been very useful in the study of spin-isospin excitations of the nucleus. In particular, the dominance of the isovector spin-flip component Vpx of the nucleon- nucleon interaction at energies above ^100 MeV acts as a filter for Gamow-Teller5130027024021018015012090M30b)IS c a l e  t.6)30T o t a l  :1057I n  c h a n r » l  163 20I n t  l i m i t s  10000202)0Sum i10 5710000 11280 12560 13840 15120 16400 17680 18960 20240XFK3002702402101806030c)1 1i n  c h a n n e l  11240I n t  l i m i t s  100002023011280 12560 13840 15120 16400 17680 18960 20240XFKFig. 68. (a) Singles proton energy spectrum 12C(p,p'); proton energy spectra in coincidencewith y-rays gated by the reals (b) and randoms (c) peaks of the time-of-flight (TOF) spectrum (d) for protons in the 12C(p,p'y) reaction.52BEADlTARfrETladder.BASEFig. 69. Top and side views of the set-up for the full (pjp'y) experiment using ten y-ray detectors.(GT) transitions at low momentum transfer. This makes it possible to study the Gamow- Teller resonance relatively free of back­ground with (p,n) and (n,p) reactions in the TRIUMF energy range.(p,n) measurements have shown that, at most, 50-60% of the model-independent sum rule, S_-S+ = 3(N-Z), is observed, where S_ is the B” decay strength which is related to the zero degree (p,n) cross section, S+ is the 8+ decay strength, related to the (n,p) cross section, and (N-Z) is the neutron excess [Goodman et al., Phys. Rev. Lett. 44^ 1755 (1980)]. The 50-60% quenching assumes that S+ == 0 for heavy nuclei because the proton- to-neutron transition in the nucleus is Pauli blocked to first order.The commissioning last fall of the (n,p) mode of the TRIUMF nucleon charge exchange facili­ty has made possible, for the first time, a purely experimental test of the sum rule and the study of GT quenching in the (n,p) direc­tion. As reported in the 1985 annual report, the 5l+Fe(n,p)51tMn reaction has been used todetermine the Gamow-Teller strength distribu­tion in 51tMn. Data at 0° and 5° for a 298 MeV neutron beam were obtained in December 1985. During the April 1986 run the angular distri­bution was completed by measurements at 2.5°, 8° and 12°. Statistics were also improved at 0° and 5°. Analysis of these data is complete and cross sections have been obtained out to 50 MeV in excitation energy. The results are shown in Fig. 70.The integrated yield up to Ex = 8 MeV is15.210.8 mb/sr. Estimating that this region is 8018% AL = 0, we obtain a GT cross section of 12.211.45 mb/sr. The ratio of the zero-degree cross section to B(GT+ ) is calculatedto be 3.6 using the DWIA and the Love-Franey interaction. The measured Gamow-Tellerstrength is therefore B(GT+) = 3.410.4. This is only about 40% of the strength predicted by shell model calculations done by Bloom and Fuller [Nucl. Phys. A440, 511 (1985)] and by Muto [Nucl. Phys. A451, 481 (1986)], B(GT+) = 9.1. This large quenching is not surprising given the severe truncation of the shell model basis used.A multipole decomposition of the spectra is being done by generating angular distribu­tions of the cross section in 2 MeV bins. It is hoped that in this way the L=0 GT strength at low excitation can be separated out more reliably, and we can search for strength at higher excitation energy.By combining these data with the 54Fe(p,n)51fCo measurements from IUCF [Rapa- port et al., Nucl. Phys. A410, 371 (1983)] B(GT~) = 7.8±1.9, we get a value of 4.4±2.0 for the sum rule. This is 73% of the pre­dicted value of 6. However, the large uncer­tainty makes this result consistent with both the observed 60% quenching in (p,n) reactions and the sum rule value of 3(N-Z). A better determination of B(GT~) must be made before definite conclusions can be drawn. This must await analysis of the 51*Fe(p,n)51tCo data taken in August at TRIUMF. During this run data were obtained at the same angles as for the 51*Fe(n,p) reaction and for the same beam energy. These data are being analysed and it is hoped that the error on B(GT”) will be reduced.53X-section X-section X-section X-Section (mb/sr/MeV)Experiment 403Pion absorption reactions 6Li, 12C(tt+,X,X2)A(G.J. Lolos, Regina)The purpose of this experiment is to investi­gate the pion absorption process in selected cases of complex nuclear systems, more spe­cifically the role of two-nucleon and multi­nucleon pion absorption processes in the presence of the nuclear medium. While pion absorption at rest or for energies less than 80 MeV appears to be dominated by the two- nucleon ir++(pn) -*■ pp reaction, evidence at higher energies points to a significant multi-nucleon role; such experimental evi­dence, however, has been extracted out of 'kinematically incomplete' experiments with­out measuring any of the four-momenta of the emitted nucleons, and as such they have come under criticism.This experiment is based on Expt. 281 and was approved for 33x12 h shifts of Mil beam time to study the pion absorption reactions on 6Li and C at T^ = 100 and 165 MeV. Improvements over the detection system used in Expt. 281 will include a compact hodoscope design and smaller MWPCs, allowing wider coverage of phase space so that the two arms will now cover a range of angles 15° < 0^,0b < 165°. This wide coverage coupled with the particle identification and the total particle energy information will allow us to test not only the older and controversial results but in addition to expand the pion absorption data base to previously uncharted dynamical regions.1 —■ 1 — 1 012 d o g .10 20  30  4-0 50E x  ( M e V )Fig. 70. 5l*Fe(n,p)5l+Mn spectra at 0°, 2.5°, 5°, 8° and 12°. The Gamow-Teller peak at low energy shows a forward peaked angular distri­bution characteristic of AL=0 transfer.54RESEARCH IN CHEMISTRY AND SOLID-STATE PHYSICSExperiment 241Temperature dependence o f reaction rate constantsfor muonium reactions in liquid phases(Y.C. Jean, Missouri-Kansas City)Chemical reaction rate constants between muonium atom (Mu) and benzene and ethylene have been measured as a function of tempera­ture in hydrocarbon liquids. These results conclude the importance of Mu dynamics in developing the current kinetics theories. The results were presented at the 4th Int. Muon Spin Rotation Conference, Uppsala, June.Solvent effect. Mu has been successfully ob­served in various hydrocarbon liquids, such as methane, ethane, propane and isopentane. The Mu formation probabilities have been found to be 15-20% for these liquids and solids in a wide range of temperatures. These liquids are ideal for kinetics studies at low temperatures. A paper based on these results has been accepted for publication in J. Phys. Chem.Mu + ethylene. Diffusion-controlled reaction rate constants between Mu and ethylene have been found significantly deviated from the Stokes-Einstein diffusion conditions in hydrocarbons. The observed activation ener­gies are found to be less than the activation energies due to viscous flow. The results are shown in Table V and Fig. 71.Mu + benzene. The chemical reaction rate constants between Mu and benzene in liquid isopentane have been measured between 136 K to 295 K. The kinetic data have been interpreted as a strong quantum tunnelling in Mu reactions. The obtained kinetic parameters for Mu addition to benzene are listed in Table VI.Table V. Kinetic parameters for reaction (Mu addition to ethylene) in various solvents (k^ = rate constant determined, Ea = activation energy determined, Evis = energy barrier to viscous flow for these solvent viscos­ities) .Solvent t P kM Ea Evis(K) (cp) (M-1 s"1) (kJ mol-1)isopentane 295 0.22 2.6*1010 4.3+0.7 10methane 110 0.12 2.1xl010 1.7+0.4 2.5water 295 1.0 1.4xl010 (15±4) 17Experiment 276Diluted magnetic semiconductors(E.J. Ansaldo, Saskatchewan)This experiment is the first application of pSR techniques to the diluted magnetic semi­conductors (DMS), of which the prototype is the system Cd(l-x)Mn(x)Te (CdMnTe). In such systems the substitution of Mn^ "*" ions for a fraction of the host cations brings about many interesting magneto-optical and elec­tronic effects, and of interest to pSR the DMS display a magnetic response resembling that of the canonical spin glasses for the concentration range 0.2<x<0.6, and an anti­ferromagnetic-like phase for 0.6<x<0.75. As yet, however, the properties and indeed the existence of a phase transition have not been fully established by the standard techniques of susceptibility, EPR and neutron scatter­ing; in particular, no direct evidence wasTable VI. Kinetic data for Mu addition to benzene in isopentane.Ea 6 .00±1.00 kJ mol-1viscosity 10.0 kJ mol-1Reaction barrier 2.1 AOscillating frequencyat transition state 400±65 cm 1Fig. 71. Plot of £n kM against 1/T for a solution of lxlO-4 M in isopentane. (Thedashed line shows AEV^S = 10 kJ/mol.)55Fig. 72. Relaxation rate X(T) of the dynamic signal for representative CdMnTe samples. Experimental beamtime and time resolution preclude the reliable determination of X 100 ps-1. The asymmetry of the signal below the transition temperature is in all cases <0.05 compared to 0.05 in the paramagnetic region.found for the spin freezing or any long-range order in the spin glass and antiferromagnetic cases, respectively, and no information on the dynamics of the spin system has been obtained. We have carried out measurements mostly in zero field to study the spin dynam­ics of CdMnTe, plus some measurements in a 4 kOe field, both in longitudinal and trans­verse geometry, to assess the internal field distribution width.The main pSR signal was found to follow a simple exponential relaxation shape in the paramagnetic region, in contrast to the root- exponential form typical of the dilute alloy spin glasses. The relaxation parameter, as shown in Fig. 72 for a few concentrations, diverges when a 'transition' temperature Tg is approached. Our measured Tg agrees in all cases with the susceptibility cusp temp­erature for the given composition x. Thus inySR the transition to the ordered phase is signalled by a critical slowing-down type of divergence in the relaxation parameter. This is similar to the behaviour observed for the dynamic component of the signal in the metal­lic spin glasses, and allows us to extract a critical exponent of the order of 1.5 (some­what x-dependent). Most of the asymmetry of the signal disappears abruptly below the transition. This effect is typical of the appearance of strong quasistatic fields in magnetic materials. The magnitude of the effect in the present case precludes us from deciding if there is a definite onset of a spin-frozen state as in the dilute spin glasses, or the extent of the remaining fluctuating dynamic component of the signal below the transition Tg. The results changed little upon application of the 4 kOe field. The transverse field measurements yielded a large and linear paramagnetic56shift, as expected for a distribution of strong internal fields in the insulators. The magnitude of the relaxation and shifts obtained indicate, by comparison with a simple estimate of the dipolar interactions, that the muons experience a sizable hyperfine interaction, perhaps with the p-electrons that mediate the exchange interaction respon­sible for the magnetic effects. This is also consistent with the comparatively large high-temperature values of the relaxation parameter, which (see Fig. 72) are an order of magnitude larger than typically obtained for x, the antiferromagnet MnF2, and which may again be attributed to the exchange narrowing limit of a sizable hyperfine inter­action. The dip in the relaxation parameter observed for x < 0.5 at 100<T<250 K is most likely due to the existence of nonmagnetic traps. As the muons became mobile (at about 100 K) in the lattice they become trapped in a site with relatively weak internal fields, i.e. away from Mn ions, and subsequently detrapped at T>200 K to experience the in­ternal fields due to the Mn ions as they hop between interstitial sites in the lattice (at a rate of about 10 ys-1). No dip is observed for x>0.5, indicating that the Mn ion concen­tration is homogeneous in those samples, so that the muon sites always have a nn or nnn Mn ion and experience the strong interaction, the hopping rate being too slow to produce a sizable motional narrowing effect.We are planning to carry out measurements in stronger external fields (>1 T) to assess the magnitude of the internal fields and their fluctuation rates.Experiment 286 Quantum diffusion o f muons (G.M. Luke, UBC)The first observation of a level-crossing resonance (LCR) was made in copper in the summer of 1984. Since that time a detailed analysis of the data has been prohibited by the lack of a full model with which to com­pare the data. Recently, a new method has been developed by Moreno Celio for theoreti­cally calculating the muon polarization function, even in cases where the muon is a part of a complicated spin system. This development has allowed exact evaluation of the polarization function as a function of a small number of physical parameters (such as the copper dipole and quadrupole frequencies)TIME ( m ic ro s e c )Fig. 73. Muon asymmetry on resonance (applied longitudinal magnetic field = 78.5 G). Sample temperature = 20 K.Using this procedure we have been able to make a preliminary fit of the quadrupole fre­quency of the nearest-neighbour copper nuclei and found it to be:cogu = -3.28±0.05 ys_1 . (1)Figure 73 shows a fit to the data on reso­nance, while Fig. 74 shows the calculated polarization function as a function of ap­plied longitudinal field.The LCR data in the temperature regime where the muon is diffusing slowly (20 K < T < 80 K) is how fully understood; however, the effects of motion require more analysis. It is clear, however, that the standard method of generat­ing dynamic polarization functions (using aFig. 74. Calculated muon polarization function for longitudinal fields 0 G < B < 120 G, for static muon.57strong collision approximation) fails near the LCR. This can be easily seen by the fact that the LCR is still quite visible at temp­eratures as high as 200 K, where the strong collision calculation predicts the effect to be completely washed out. Further work in­volving, amongst other things, correlations between adjacent muon sites is in progress in order to understand the high temperature data.Experiment 296Gas phase muonium addition reaction kinetics(D. Garner, TRIUMF)Results of the preliminary data analysis for this experiment were presented in last year's annual report. Data analysis is now complete and was presented at the 9th International Symposium on Gas Kinetics at Bordeaux in July. Several journal publications are now in preparation.A. Wagner and R. Duchovic at Argonne have calculated the muonium addition rate con­stants for acetylene using their new ab initio potential in an attempt to sort out the failure of theory to agree with the ex­perimental data for the analogous H atom reactions. Although an adjustment of the potential yields remarkably good agreement with the Mu rate constants at all tempera­tures, the potential does not simultaneously reproduce the H atom data. There is now some question as to whether any of the Mu or H atom experiments are actually in the high pressure limit as claimed. The Argonne group presented a joint paper on the experiment and theory of this system at the Combustion Conference in Munich in August.This question has motivated us to work out what effect being below the high pressure limit would have on the raw data of the ex­periment. R.E. Turner of UBC Chemistry worked out the formalism, which shows that the ex­periments conducted so far are measuring what the experimenters have claimed. We are now working in collaboration with the Argonne group to calculate under what conditions, if any, will a pressure dependence be experi­mentally detectable. This will either lead to a new experimental proposal, or the ques­tion of whether the experiment is in the high pressure limit will be resolved from the cal­culations. In any case we are working toward a joint paper with the Argonne theorists which should help settle some of the out­standing questions on the H + acetyleneaddition reaction, which is a very important reaction in the field of combustion.Experiment 339Kinetic isotope effects in the Mu+H? and Mu+D?reactions (I.D. Reid, UBC)Experiment 339 was proposed to study the ef­fects of isotopic substitution in the basic chemical reaction H + H2 H2 + H. This re­action is the only one for which exists an accurate ab initio potential energy surface, known as the LSTH surface, and there has been considerable interest in calculating reaction rates on this surface. The theoretical re­sults can be tested by comparison with ex­periment, especially those experiments where one or more of the H atoms is replaced by an isotope (e.g. T, D or Mu) - the Born-Oppen- heimer approximation implies that the LSTH surface is valid for all such substitutions. In this experiment we have remeasured muonium reaction rates with hydrogen and deuterium (e.g. Mu + H2 MuH + H) to much higher pre­cision and over a wider temperature range than previous experiments [Garner et al., Chem. Phys. Lett. 121 , 80 (1985)]. The new results, over the range 600-843 K in D2 and 473-843 K in H2, are sufficiently precise to discriminate between competing theoretical treatments and are in excellent agreement with the most accurate theory.The experiment had to cope with unusual re­quirements, due to the slowness of the reac­tions being studied. To obtain muonium spin relaxation (MSR) rates which were measurable during the 2.2 us muon lifetime, we needed very high pressures of hydrogen and deuteri­um, up to 1.4 MPa, as well as the tempera­tures noted above. A major constraint on our reaction vessel was the provision of a low mass thickness entrance window for the sur­face muons, coupled with the safety require­ment of proof testing at twice working pres­sure and full working temperature. We used a 0.05 mm Inconel foil formed into a 2 cm dome for the window and it passed all tests with flying colours. The rest of the reaction ves­sel, 25 cm diameter and 64 cm long, was welded out of stainless steel, since the MSR technique requires nonmagnetic materials. The vessel and its electrical heater were con­tained within a series of heat shields and a vacuum jacket to provide thermal insulation.The measurements were conducted on the M15 surface muon channel, taking advantage of the58small beam spot available on this beam line. Being able to use a small collimator and still obtain high incident muon rates was a major factor in the design of the vessel, since it allowed the use of the small en­trance window. We measured muonium relaxation rates at several reactant pressures at each temperature and from the variation in relaxa­tion with concentration then calculated the reaction rate constants. These rate constants are plotted against temperature in the Arrhenius diagram (Fig. 75). Also included in the figure are two current reaction theory treatments: the 3-dimensional quantum coupled states (CS) calculations of Schatz [J. Chem. Phys. 83, 3441 (1985)] and variational trans- ition-¥tate theory (ICVT/LAG) calculations due to Garrett and Truhlar [J. Chem. Phys. 81, 309 (1984)]. It is immediately apparent from the figure that the 'exact' CS calcula­tions do indeed provide a good description of the reaction, while the ICVT/LAG treatment is significantly worse in its predictions. Our results are thus extremely important in test­ing reaction rate theory as they now provide sufficiently accurate data to allow the theo­rists to more rigorously evaluate their methods, the LSTH surface, and the validity of the Born-Oppenheimer approximation.Experiment 340Muon molecular ions and ion-molecule reactions (D.G. Fleming, UBC)This experiment is concerned with measure­ments of the ion-molecule reactivity of the y+ molecular ion [Mmu+ ]* in rare gases (M = He, Ne, Ar) with a variety of reactants 'X', according to the (simplified) reaction scheme1000 833Temperature (K) 714 625 556 500 455[Mmu+ ]*+X[Mmu"*"]Mu + X+ + Ne XMu+ + NeHere kc, kt and kq are bimolecular rate con­stants for charge exchange, muon transfer and 'quenching' (by M), respectively. Central to the interpretation of these reactions (see 1985 annual report) is the presence of rovi- brational excited states [Mmu+ ] , found during the initial u+ capture process. This is necessary in order to guarantee that the charge exchange channel (kc) is exothermic, as required for these generally fast reac­tions at thermal energies. For reactions with X, charge exchange is generally the only mechanism available for dephasing the muon spin and hence giving rise to a relaxation ofno£oEE0*1oco-*-»wcoCJocc1000/Temperature (K)Fig. 75. Experimental measurements and theo­retical predictions of the reaction rate constants for the reactions of muonium atoms with hydrogen and deuterium molecules.the (diamagnetic) ySR signal. In the partic­ular case of NO the proton transfer channel (kt) also causes dephasing since the product ion (like Mu itself) is paramagnetic.In the past year additional data have been obtained on the reactivity of [ArMu+ ] as well as on the effects of ternary mixtures on the relaxation process.In the case of the [ArMu+ ] ion we have studied reactions with X = Xe, 02, NO, NH3, CH^, CH3N02, TMS (tetramethylsilane) and TEA (triethylamine); of these, relaxation was ob­served only for NO and TEA with rate con­stants kex = (3.511.5) and (7±5)xl0-l6m3S-1, respectively. Both of these values are with­in a factor of two of the theoretical (ADO) prediction and as such are consistent with other cases previously discussed (1985 annual report).More significantly, these are the only two cases where reaction of [ArMu+ ] with X is clearly exothermic from the ground state of the ion, strongly suggesting that kq is59sufficiently large that any [ArMu ] ini­tially formed is efficiently quenched by collisions with Ar during the slowing-down process. Estimates based on elastic modera­tion for the slowing-down time ts give ~10 collisions for ts ~ 20 ns. This situation is very comparable to that existing in low pressure ICR studies, where the relative efficiency in quenching of the corresponding [ArH+ ] ion by Ar is well established [Smith and Futrell, Int. J. Mass. Spec. & Ion Phys. 20, 33, 43, 59, 71 (1976)]. The magnitude of kq in our studies is estimated to be kq ~ 5xl0-18 m3s-1, in contrast to those for the*4“  ^ | ^[HeMu ] and [NeMu-1"] ions, which are at least an order of magnitude less. These values are consistent with those reported elsewhere for comparative studies of quench­ing of a variety of molecular ions by He, Ne and Ar [Durup-Ferguson et al., J. Chem. Phys. 81, 2657 (1984); Dotan et al., Chem. Phys. Lett. 121, 38 (1985)].ISome preliminary results on the effects of ternary mixtures on the relaxation of the pSR signal have proven interesting and are worthy of further study (Fig. 76). The basic moti­vation for these studies is to learn about the relative importance of quenching vs. reactive collisions in the ion molecule chem­istry of the 'muonated' rare gases. Specifi­cally, in studies of [HeMu+ ] and [NeMu+] there is n£ observable 'fast' (charge ex­change) relaxation with X = CH^ , C2Hg or H20, cases which are just as exothermic as others that we have studied which do give rise to relaxation (e.g., Xe, N20, CF^). This sug­gests that cross sections for thermal charge exchange are much reduced for these cases, a contention supported by the relatively weak dependence of amplitude on [X], or, concomi­tantly, that the muon transfer channel is enhanced. By studying ternary mixtures, e.g., [NeMu] /NH3/X, where one component giving rise to charge exchange and hence a fast relaxation is fixed (NH3), while the other giving rise to nonreactive (X = H20) muon transfer collision is varied, a value for the sum (kt+kq') can be determined. Here kq' means quenching with reactant X. This latter feature is an important component of more complicated kinetic schemes that we are currently discussing. In Fig. 76 relaxation for fixed [NH3 ] have been measured for X = H20 (upper) and X = C2Hg (lower). The slopes correspond to rate constants of kexp = (28±6) xIq-16 an(} (6±3)*10~15 m3 s-3, respectively. The result for water is in good agreement with proton transfer studies of rare gas ions but that for C2Hg seems too low [Villinger etTernary mixtures: X +  nm, + Ne[X ] (10“/ cm3)Fig. 76. pSR relaxation rates for the tern­ary mixtures [NeMu+ ]*/NH3/H20 (upper) and [NeMu+ ]*/NH3/C2Hg (lower), where the con­centration of the third component (X = H20, C2Hg) is varied. The slopes give the experi­mental rate constants kexp = (28±6)xl0~16 and (6±3)xl0-16 m3 s-1 for H20 and C2Hg, respec­tively .al., J. Chem. Phys. 7h_, 3529 (1982); Mackay et al., Can. J .  Chem. 59_, 1771 (1981)].Further studies are in progress, including ternary mixtures where both components are reactive to charge exchange, which will pro­vide additional tests of our kinetic models.Experiment 347Spin dynamics in crystalline and amorphous DyAg (D. Noakes, UBC)Most of the work this year has been analysis of data obtained at TRIUMF in calendar 1985 (see Annual Report Scientific Activities 1985, p. 63). Most interesting as pSR phe- nomonology is the relaxation in the antifer- romagnetically ordered state below T^ = 60 K in the crystalline sample, which at lowest temperatures is clearly of the static Lorent- zian Kubo-Toyabe shape in zero field and in longitudinal field 'decoupling' (Fig. 77). This shape is usually associated with dilute disordered 'spin glass' alloys, and is prob­ably due to disorder in the antiferromagnetic state with the muons sitting at a high symme­try site at which the perfectly ordered state field would be exactly zero. However, there is complicated behaviour at higher tempera­tures (still within the antiferromagnetic state). In particular, in longitudinal field60Q)EE>,X I0)oCDoaTIME (m icrosec)Fig. 77. MSR asymmetry spectra of crystalline DyAg at 18 K in zero and longitudinal fields as indicated. Solid lines are least squares fits of the static Lorentzian Kubo-Toyabe re­laxation function for the fields indicated.T im e (m ic ro s e c )Fig. 78. ySR asymmetry spectra of crystalline DyAg at 32 K in zero and longitudinal fields as indicated. Note that the 950 G and 3800 G spectra are nearly indistinguishable.there is partial decoupling in fields up to about 1 kG, beyond which higher field (to a maximum available at the time of 3.8 kG) had no effect (Fig. 78). It is not clear at this time what combination of possible microscopic effects (such as static defects/impurities, domain walls and spin waves) will reproduce the observed phenomena, but analysis is still in progress. Preliminary results of this experiment will be published in the Proceed­ings of the 4th International Conference on ySR [Hyp. Int. (in press)].We now are able to do ySR at TRIUMF in up to 30 kG applied magnetic field, and we plan to extend the measurement of longitudinal field ySR in DyAg samples to these higher fields in the near future.Experiment 349Thermally activated muonium formation in BaF?: and AZ 2O3 (S.R. Kreitzman, UBC; E. Ansaldo,Saskatchewan)The focus of this experiment, i.e. the site location of the diamagnetic y+ fraction in alkali-halides and other canonical insulators has been broadened into the related field of the radiolysis reactions associated with this fraction. Impetus for this new direction is the first direct observations of thermal Mu formation in Af.203 and BaF2 • The signatureof the reaction in both materials is the very rapid decay of the transverse field (TF) pol­arization compared with depolarization rates that would be typical of any possible dipolar interactions. This is illustrated in Fig. 79 where TF data for A.SL 03 are shown at 10 K and 18 K. Comparable behaviour in BaF2 occurs near room temperature between 240-260 K.A second significant feature found in both materials is the nonexponential decay enve­lopes which each displays. We interpret this to indicate time-dependent rates of Mu forma­tion which are expected to accompany a therm­al formation process. The form of the decay envelope, however, is very different for BaF2 than for the 18 K A5.„03 data above. Specifi­cally, even at the highest temperature mea­sured, 325 K, there remains a time-indepen­dent muon fraction which does not convert to Mu. Combining these data with previous work, which indicate that prompt fractions of Mu and y+ exist [Kiefl et al., Phys. Rev. Lett. 53, 90 (1984); Minaichev et al., Zh. Eksp.Teor. Fiz. 62_, 14 (1972); Nishiyama et al., Phys. Lett. Ill, 369 (1985); Brewer et al., submitted to Phys. Rev. B. Rapid Commun.] and that Mu <-> p+ reactions abound, will help delineate the qualitative features of the electron transport, charge capture and/or site changes which surely accompany the Mu formation that we observe.6 1TIME (Microsec)Fig. 79. Low temperature TF data in a single crystal of AL203. The 18 K data show a mean relaxation ratfe of 10 us-1 indicating Mu for­mation at this rate. Note the very rapid in­crease of the formation rate as a function of temperature.Experiments 358, 398 fxLCR spectroscopy o f free radicals (R. Kief I, TRIUMF; RW. Percival, Simon Fraser)Muonium-substituted free radicals were first detected less than a decade ago, and muchuseful information has been gained on their formation, structure and reactions. However, the scope of the investigations has beenlimited by the method of detection, time- differential transverse-field pSR. Experi­ments are almost always carried out in high magnetic fields, where only two muon preces­sion frequencies are observed. The muon-electron hyperfine coupling is readily obtained as the difference between the two precession frequencies, but information on the hyperfine coupling of the other nuclei is lost. Furthermore, the problem of spin- dephasing restricts detection to those radi­cals formed very rapidly (within 100 ps or so, if formed from muonium), so that onlypure samples or extremely high concentrations of the radical precursor may be studied. These restrictions are lifted in the muon level-crossing resonance (yLCR) technique, in which non-muon spin transitions are probed by monitoring the muon polarization as the longitudinal magnetic field is swept.0.04- o  0 4  -------- 1-------- 1-------- 1-------- 1-------- 1-------- 1-------- 1-------- 1_____1_____20 22 24 26 28 30MAGNETIC FIELD  (kG)Fig. 80. yLCR spectrum of CgHgMu in benzene.The yLCR effect was first demonstrated at TRIUMF, in Expt. 286 (Kreitzman et al., Phys. Rev. Lett. 56, 181, 1986), and was then ex­tended to paramagnetic systems in Expt. 358 (Kiefl et al., Phys. Rev. A 34_, 681 (1986). The free radicals side of Expt. 358 has been continued as Expt. 398, which was approved in July 1986.To date, LCR signals have been detected for free radicals formed from benzene, hexadeute- robenzene, hexafluorobenzene, 1,2,3-trihy- droxybenzene, furan, ethene, 2-methylpropene and 2,4-dimethyl-2-butene. An example of a yLCR spectrum is given in Fig. 80. Separate resonances can be seen for each magnetically distinct proton in the cyclohexadienyl radi­cal CgHgMu formed by addition of muonium to benzene. Hyperfine coupling constants, calculated from the resonance positions, are collected in Table VII and compared with ESR data for the corresponding H radicals.Comparison of the hyperfine couplings mea­sured by yLCR and ESR shows that replacement of a methylene H by Mu does not perturb the it spin density distribution in the ring. This suggests that the isotope effect on the CH2 coupling (the higher muon and lower proton coupling in CHMu) is due to a geometrical effect, namely out-of-plane libration of the CH2 (CHMu ) group. The magnitude and particu­larly the temperature dependence of such isotope effects gives valuable information on intramolecular motion. Of particular interest is rotation about carbon-carbon bonds in acyclic alkyl radicals. Beta proton hyper­fine coupling constants in alkyl radicals obey the empirical relation:A = Bq + B1<cos20> ,62Table VII. Comparison of hyperfine frequencies in muonium cyclohexadienyls and analogous H radicals.Radical Ay /MHza NucleusA-l/MHz Mu radicalAi/MHz H radicalC6H6Mu 161.61(1) H(6 ) 126.11 (4) 134.61(6)/c6h 7 H(1,5) H(2,4) H(3)-25.14 (4) 7.47 (4) -36.19 (4)25.22(6)7.54(6)36.83(6)C6D6Mu 163.27(1) H(6 ) 136/c6d 6h 0(6 ) 0(1,5) D(2,4) D(3 )19.09(12)-3.92(12)-5.59(12)22.2C6F6Mu 63.04(1) H(6 ) 54.1(5)/c6f 6h F(6) F(1,5) F(2,4) F(3)357.24 (4) 67.20 (4) -16.27 (4) 104.62 (4)354(4)67.5(7)16(2)105(1)aA^ = 0.31413 Aywhere Bfl and B1 are constants (B1 »  BQ ) and 0 is the dihedral angle between the axis of the p-orbital at Ca (which contains the unpaired electron) and the Cg-H bond. The observed coupling is a weighted average over conformations with different 8. With yLCR it is now possible to measure the hyperfine coupling for both a proton and a muon substi­tuted onto the same carbon. Any change in B3 would affect both substituents, whereas a preferred conformation which enhances the muon coupling would have the opposite effect on the proton coupling. Some results on substituted and unsubstituted methyl groups in the radicals CH2CH2Mu, (CH3)2CCH2Mu, and (CH3)2CCMu (CH3)2 are given in Table VIII.The presence of Mu in the radical has only a minor effect on unsubstituted methyl groups, but breaks the symmetry of the methyl group substituted, so that the hyperfine frequen­cies are strongly temperature dependent, and a preferred conformation is taken up in the low-temperature limit. The conformational effect can be set aside by considering the average coupling for the CH2Mu group, definedas A - (A^+2Ap)/3. [Ay represents the muon frequency reduced by the ratio of proton and muon magnetic moments.] In the ethyl radical A is close to the ESR result, but not so in t-butyl! Detailed studies of temperature de­pendence are necessary to explore this matter further; the t-butyl study is well under way (see Figs. 81 and 82). As can be seen from Fig. 82, A(CH3) is roughly constant, but A(CH2) rises and A(Mu) falls with tempera­ture. The mean for the CH2Mu group is nearly independent of temperature, but there is a significant deviation, probably caused by a departure from planarity at the radical centre. Further discussion is deferreduntil the final analysis is complete..O lO .005 0.0—  .005 -.010.o io.0050.0—  .005 —-O IO.005 0.0 —  .005 —-O IOFig. 81. yLCR spectrum of (CH3)2CCH2Mu at various temperatures.291 K -li^ .WV'lW*WrT'|TTlVtil'TlTT4Al*Mi1.2 1.3 1. 4- 1.5M A G N E T IC  F I E L D  ( T )Table VIII. Hyperfine frequencies in muonium-substituted alkyl radicals.Radical Temp/K Ap(CH3)a Ap(CH2Mu) A^(CH2Mu ) A(CH2Mu )A (CH3) by ESRb(ch3)2ccmu(c h3)2 300 63.7 64.2(c h3)2cch2mu 289 63.0 55.4 91.6 67.5 63.7c h2ch2mu 300 62.0 103.6 75.9 75.3aall frequencies in MHz ^analogous H radical63Hyperfine Couplings (MHz) in t—butylTemperature /  KFig. 82. Temperature dependence of hyperfine frequencies in (CH3)2CCH2Mu.Experiment 361Muonium in water(PM  Percival, Simon Fraser)When positive muons stop in water roughly 60% are incorporated in diamagnetic compounds, and the remainder form muonium. The two fractions can be distinguished by their muon spin rotation (pSR) signals, but the signal amplitudes indicate that part of the initial muon spin polarization is missing. It is now generally accepted that the depolarization occurs as a result of encounters between muonium atoms and hydrated electrons formed at the end of the muon track. These encoun­ters take place roughly 1 ns after muon ther- malization. However, simple considerations of spin suggest that only one-quarter of all encounters should result in nonreactive spin exchange and that another quarter should re­sult in reaction, presumably (by analogy withH):Mu + eaq *  MuH + OH* .In the transverse magnetic fields used in pSR distribution of reaction times for individual muonium atoms results in loss of phase coher­ence of the muon spins in the diamagnetic product, i.e. there is loss of signal for the product MuH. However, a careful examination of the field dependence of the diamagnetic signal revealed a small increase in amplitude at low fields, consistent with formation of MuH from Mu at a similar rate to the muoniumdepolarization process. The effect of varia­tion of temperature, pressure and concentra­tion of paramagnetic ions have all been in­vestigated. The paramagnetic ions quench the field dependence by depolarizing Mu before encounter with e™. The overall level of dia­magnetic signal increases with temperature and pressure, but the field variation is un­affected. These results suggest that there are two distinct contributors to the diamag­netic signal in water: MuOH formed at veryshort times (< lps), and MuH formed by Mu -*• e^q encounters.Experiment 362High pressure muon spin resonance in liquids(PM. Percival, Simon Fraser)A high pressure cell has been developed to enable muon spin rotation studies of liquid samples under pressures up to 3000 atm. A muon stop rate of 14,000 muons/s is achieved using the standard M20A backward muon beam tune with a final collimator of 10 mm diam­eter in front of the 12 mm diameter cell window. Under these conditions 90±3% of the muons stop in the sample; only 10% are scattered into the cell walls.As an initial test of the system water was used as the target material, and muonium and diamagnetic muon precession signals were recorded at various pressures up to 2000 atm. The diamagnetic signal amplitude was found to increase with pressure, while the muonium amplitude decreased. These results will be incorporated with other measurements on muonium in water taken under Expt. 361.After the initial feasibility studies beam time was devoted to the study of the pressure dependence of muonium kinetics. To date, muonium decay rates have been measured in solutions of sodium nitrate and potassium permanganate under pressures from 1-2000 atm. The rate for Mu + NO^ was found to increase with pressure, indicating a positive volume of activation,AV* = -RT(81nk/3P)T ,in contrast to the Mu + MnO^ rate which de­creased with pressure. The signs are in accord with H atom data for aqueous reac­tions, from which it is known that diffusion- controlled reactions have positive AV* and activated reactions negative AV^. Quanti­tatively, we find AV* = -7.1± 1.5 cm3 mol*1 for the reaction with nitrate, somewhat64larger than the published value of 5 cm3 mol-1 for H. This is consistent with a greater molar volume for Mu, as predicted by molecular dynamics calculations carried out by Klein et al. at NRC Ottawa.Experiment 367Resolved nuclear hyperfine structure ofanomalous muonium in semiconductors(R. Kiefl, TRIUMF)The electronic structure of a muonium defect centre in a semiconductor is identical to that of hydrogen, except for small zero point motion effects resulting from the fact that the muon has only 1/9th the proton mass. Al­though conventional resonance techniques such as electron paramagnetic resonance and elec­tron nuclear double resonance have yet to detect hydrogen in a semiconductor, two quite different kinds of muonium defect centres have been observed via the technique of muon spin precession. Both normal muonium (Mu), with its large isotropic muon hyperfine (yhf) interaction, and anomalous muonium (Mu ), with its small highly anisotropic yhf inter­action, have been observed in diamond, sili­con, germanium and most recently in GaAs and GaP.One of the most intriguing problems in the study of muonium defect centres has been the mystery concerning (Mu ), i.e. where is the muon situated and what is the charge and electronic state? Until now it has only been possible to measure the muon hyperfine inter­action, which by itself does not provide a clear picture of the Mu centre. Resolved nuclear hyperfine structure would provide important additional information on the spin density distribution near the muon, which could be used to test current theoretical models for Mu* and the associated methods for calculating its electronic structure.In the past year we have used the technique of level-crossing resonance (LCR) to resolve the nuclear hyperfine (nhf) structure of the Mu* centre in GaAs and GaP. A few of the LCR spectra taken on GaAs are shown in Fig. 83. These spectra are considerably more complicated than LCR spectra for muonium- subsituted free radicals (see Expts. 358 and 398) because of anisotropy in the yhf and nhf tensors and nuclear quadrupole (neq) interac­tions for nuclei with spin greater than 1/2. From the more prominent LCR lines above about 0.8 T we have extracted nhf parameters for the nearest neighbour 75As(31P) on the Mu0.040.020.0I<1 0.04+<0.020.0- 0.020Fig. 83. Level crossing resonance spectra in GaAs for Hg applied along the <100> and <111> crystallographic directions. The resonances occur at specific magnetic fields where the muon spin transition frequency is matched to that of a neighbouring nucleus. On resonance there is a transfer of polarization from the muon to the nucleus. The 73As resonances appear as triplets which are split by the nuclear electric quadrupole interaction. The lines are labelled by the angle between the symmetry axis and the applied field and the sign of z-component of electron spin.symmetry axis in GaAs (GaP) (see Table IX). The results indicate that about 40% of the unpaired electron is centred on this nucleus, in what is primarily a p-type atomic orbital. Along certain orientations [see, for example, Fig. 83(b)] we have also observed many other LCRs at lower fields (0.3-0.8 T), which are attributed to more distant nuclei with nearly isotropic nhf parameters in the range 30-100 MHz.Table IX. Hyperfine and quadrupole parameter for nearest neighbour group V nuclei on the <111> symmetry axis of Mu* in GaAs and GaP.Nucleus A*A l Ali Qi (MHz) (MHz) (MHz)73As in GaAs 562.5(4) 129.0(1) 18.8(2)31P in GaP 620.2(4) 249.7(1)75As( ’(54.7°75Asl+)(54.7°)BII < I00>75As(- V )BII <  III>0 .5  1.0 1.5 2 .0 2 .5 3.MAGNETIC FIELD (T)65So far the LCR spectra support the recent bond centre model for Mu* in which the muon is located in the centre of the Ga-As bond, with the unpaired electron in a nonbonding Ga-As molecular orbital. The Ga LCRs, which are predicted to occur in higher magnetic fields, should firmly test this model of Mu in semiconductors.Experiment 371 Muonium in micelles (D.C. Walker, UBC)This pSR experiment is designed to explore and exploit the effects of added micelles on the reaction of muonium towards solutes in water. The results are proving to be most interesting. When the solute happens to be localized within the added micelle then the reaction rate increases - sometimes enormous­ly (60,000-fold, for instance). When the solute remains in the aqueous phase the reac­tion can be inhibited slightly by addition of micelle.The major points are summarized by the re­sults reported in Table X. Styrene is a solute which reacts with muonium on first collision in pure water, so its 9-foldmicelle-induced enhancement probably reflects the larger effective encounter probability when solubilized. The N03~ data, on the other hand, indicate that muonium itself is not strongly localized by unoccupied micelles, perhaps only slowed down on passing through. But for 2-propanol there is a 60,000-fold en­hancement. This is a solute which reacts quite slowly with muonium in water (1.3 x 106 M-1s-1); but when the microenvironment is changed to that of amphiphilic surfactants the reaction probability per encounter approaches unity.2-propanol has shown the largest rate en­hancement so far, but even slower intrinsic reactions now need to be looked at. If this is a general phenomenon then muonium may be observed to react with any additive merely by solubilizing it in a micelle.CTAB and SDDS micelles have been used in chemistry to mimic the role of membranes in biology by creating phase-separations across whose boundaries only certain species pass. Therefore, these preliminary results suggest the possibility of muonium being used to elucidate those kinetic parameters of elemen­tary reactions which control biophysical processes mediated by membranes.Table X. Effect of added micelles on reactions of muonium with solutes in water.solute(conc/M)a Micelle'3 X-Xo/106s-1 k^/M^s-10 Enhancement''factorstyrene (2.17xl0_l+)(4.5xl0-5)N03- (l.OxlO-3)2-propanol (0.70)(2.2xl0-5)- 2.03 O.93xlOi0CTAB 3.5 8xl010 9- 1.70 1.7x109SDDS 0.75 0.75xl09 0.4- 0.87 1.3x109CTAB 1.70 7.7xl010 60,000aSolute concentrations have to be varied until X falls into the limited observable range for a ySR experiment of 0.3 5 ps_1.'’CTAB is cetyltrimethylammonium bromide, SDDS is sodium dodecylsulphate. Cone, comparable to that of the solute. ckj4 calculated as (X-X0)/[S]q> where [S]-p is the number of moles of solute added per litre of total solution.''Ratio of k^ (with micelle)/k^ (in pure water).66Experiment 385Muon polarization in Xe up to 5 atm(M. Senba, UBC)The muon spin polarization has been measured in Xe as a function of pressure up to 6.1 atm. Even at the highest pressure inves­tigated the total muon polarization is not fully recovered. An attempt is being made to understand the muon polarization in the gas phase in terms of the atomic cross sections of the collision involved in the slowing down of muons.The amplitudes of ySR signals from muonium and muons in diamagnetic environments have been measured in Xe with transverse low (7 G) and high (300 G) fields, respectively. The total muon spin polarization (Pt) is re­lated to these measured amplitudes,pt = pMu + PD where PMu = 2A^U/(A^-A^)and PD = (Ad-Aw )/(A^-Aw)where A^ is the initial muon polarization measured in an aluminum plate placed in the evacuated target vessel, A^ is the wall contribution, and the factor of two accounts for the fact that we do not observe so-called singlet muonium. At all the pressures inves­tigated no observable signal from muons in diamagnetic environments in the gas phase has been detected (Ap> = 0.01±0.01). The total polarization is shown in Fig. 84 as a func­tion of pressure. Only 80% of the initial polarization is accounted for at the highest pressure (6.1 atm), indicating the possibili­ty of a true missing fraction [Turner et al., Int. J. of Quantum Chem. 29_, 1493 (1986);Turner and Senba, J. Chem. Phys. 84^ , 3776(1986)] in the gas phase. It should be noted that similar behaviours have been observed in some heavy organic vapours with low ioniza­tion potentials [Arseneau et al., J. Phys. Chem. 88, 3688 (1984)].The usual assumption is that the muonium formation during charge exchange cycles is the only cause of the muon spin depolariza­tion in the gas phase. If muonium is formed at energy E, the average time before the next electron loss collision is expressed asT(E) = l/[nV(E)aL(E)] (1)where n is the number density of the moderat­or, V(E) is the velocity of the muonium atom at E, and ap,(E) is the electron loss cross section of muonium at E. The polarization isPressure (atm)Fig. 84. Total polarization in Xe.calculated by,P = n cos(WT(Ek)) (2)kwhere W is the muonium hyperfine frequency and the product is calculated over all sing­let muonium formations. Therefore, we expect P+l at sufficiently high pressures.In order to understand the charge exchange regime in the gas phase a formalism has been developed to describe the slowing down mech­anism of charged and neutral particles in terms of the atomic cross sections of the processes involved [Senba (in preparation for publication)]. It has been found that pro­cesses competing with charge exchange colli­sions, especially elastic scatterings, play a crucial role in determining the fraction of muons thermalized as muonium and the muon spin polarization retained down to thermal energy. At the time of the next electron loss collision the muonium atom has lost some energy due to these competing processes. Therefore, in order to calculate the average muonium residence time T(E) with Eq. (1), the quantity V(E)CTk(E) should be evaluated at energies less than E. Furthermore, the strong competition of processes other than charge exchange reduces the number of times that muonium forms, causing a change in the polarization [Eq. (2)]. Detailed calcula­tions based on this formalism reveal (i) as expected, Xe is an inefficient moderator of muons and it takes about 70 charge exchange cycles to slow down a muon from 1 keV to thermal energy; while (ii) He is a much more67efficient moderator of muons than Xe, requir­ing only about 15 cycles to effect the same moderation.By mixing He in Xe we can change the number of charge exchange cycles in a systematic way. Very preliminary measurements have been done on mixtures of Xe and He, showing a sub­stantial exhancement of polarization. These results are summarized in Table XI.Table XI. Enhancement of muonium pol­arization in Xe on addition of He.Xe(Torr) He(Torr) P(muonium)380 0 0.50380 400 0.723400 0 0.803400 400 0.82Experiment 391 Muon spin rotation studies o f supported metal catalysts (R.F. Marzke, Arizona State)Five shifts of beam time were allocated to Expt. 391 during 1986, for muonium spin rota­tion (MSR) studies of the materials used in heterogeneous catalysts as supports for small transition or noble metal particles, which are the primary working elements of these catalysts. The supporting materials play an extremely important role and must meet two criteria: (1) high surface area (>100 m2/h),and (2) chemical and physical stability under chemical reaction conditions. A commonly used support is finely divided silica (Si02), which is highly stable and is produced in commercial amounts at various levels of puri­ty. One of these supports, called Cab-0-Sil EH-5, has been the subject of study by MSR at TRIUMF, primarily in Expt. 191 (now con­cluded) . An extension of this study was made in Expt. 241 (also concluded) to EH-5 con­taining varying amounts of small platinum particles, i.e. to actual working catalysts. In the latter experiment clear evidence was seen for interactions between muonium, whose chemistry is that of an isotope of the ex­tremely reactive species atomic H, and the silica-supported Pt particles, whose surfaces are typically covered with chemisorbed layers of reactants such as oxygen.The interpretation of Expt. 241, and of other studies involving molecular species physi- sorbed onto the surface of silica, has, how­ever, been criticized. In the supporting material EH-5 it was found in Expt. 191 that the muonium relaxation rate as a function of temperature showed a broad, prominent peak near 24 K. This makes it difficult in study­ing supported catalysts, for example, to dis­tinguish between behaviour of the Mu relaxa­tion rate which is related mainly to the support on the one hand and behaviour due to the presence of adsorbates or metal clusters on the other. Clearly, it would be preferable to use for a variety of experiments of chemi­cal and physical interest a support which causes little or no relaxation of Mu overmost of the temperature range between 5 and ~350 K. Such a material could be described as having a magnetically 'clean' surface (where Mu spends much of its time after being ejected from the interiors of the silica grains following its formation). This termi­nology is introduced by analogy to that ofsurface science, where cleanliness of surfaces is generally monitored via Auger electron spectra.For Expt. 391 it was therefore proposed in July that a search be conducted for suchmagnetically clean supports, and two possible lines of investigation were suggested. The first arose from the surprising results of Expt. 241, which indicated that lightly plat­inum-loaded (0.001% to 0.01% Pt by weight) Cab-0-Sil EH-5 was a suitable material, the effect of the small amount of Pt being anunexpected, large reduction in the Mu relaxa­tion rate above 20 K. The second suggestion was to try another supplier of silica, in hopes of finding a material whose intrinsic transition metal impurity content (mostly iron) was well below the ~6 ppm found for EH-5. The reason for seeking samples with low iron content was that this impurity was believed to play a leading role in producing the large Mu relaxation peak seen in this support in Expt. 191.During this year only the second approach was tested, there having been insufficient time to initiate further studies of the lightly Pt-loaded EH-5 samples used in Expt. 241. A candidate silica was obtained through the courtesy of EM Industries, whose transition metal impurity was specified to lie in the parts per billion (ppb) range. The results plotted in Fig. 85 show the Mu relaxation rate as a function of temperature for this material in two different states of surface preparation: (1) 'bare', i.e. after pumpingon the sample at 295 K (open squares), and (2) after a small amount of air had been6 8Tem peratureFig. 85. Muonium relaxation rate in 7 G transverse field as a function of temperature in EM optical fibre-grade silica powder.admitted to the sample (filled squares). The dramatic differences seen in the two cases are believed due to the strong paramagnetism of dioxygen physisorbed on the silica sur­face. The extremely low relaxation rate ob­served over the entire temperature range above 20 K for the bare material is evidence that it is magnetically very clean indeed, and lends strength to the argument that the peak in EH-5 is actually due to the spin-flip scattering of Mu by iron (presumably Fe3"*').Experiment 402Hyperfine structure o f muon and muonium defect centres in magnetically ordered fluorides (R. Kiefl, TRIUMF)A number of experiments have been performed on the antiferromagnetic insulators MnF2, CoF2 and FeF2. In MnF2 muon-nuclear level- crossing resonances (LCRs) have been observed (see Fig. 86), where the spin transition fre­quency of an interstitial muon is matched to that of the nearest-neighbour 19F nuclei, resulting in a resonant transfer of muon pol­arization to the nuclei. This double-reso­L o n g i t u d i n a l  M a g n e t i c  F ie l d  (T )Fig. 86. The muon-^3F level-crossing spectra in MnF2 as a function of temperature below the N€el temperature (67.3 K). The upper resonance at 65 K is off the scale at lower temperatures.nance technique allows the measurement of the local magnetic field on the neighbouring 19F nuclei. The shift in the measured local field relative to the previous NMR measure­ments is a result of the disturbing influence of the muon. Additionally, a second resonance was observed when the applied field cancels the longitudinal component of the local field at the muon. The external field at which the resonances occurred was observed to approxi­mately scale with the sublattice magnetiza­tion as a function of temperature.In CoF2 and FeF2 zero-field experiments were performed using a special high-timing resolu­tion spectrometer, confirming the presence of a high-field centre (already observed in MnF2). This centre is thought to be muonium (p+e~).69THEORETICAL PROGRAMIntroductionThe Theory group at TRIUMF exists to provide a focus for theoretical research and a group of active researchers who are interested in the physics relevant to the present experi­mental program and the proposed KAON factory. As befits a laboratory with as wide a range of activities as TRIUMF, the interests of the Theory group are quite broad, including ySR, proton-nucleus interactions, meson-nucleus interactions, meson-photon-baryon reactions, nucleon structure, lattice gauge calcula­tions, grand unification, weak interactions, and even astrophysics. Part of this research involves working directly with the experimen­talist on particular experiments; part with more general physical understanding; and part with formal theoretical developments. This breadth allows the Theory group to be of benefit to TRIUMF in both the short and long term.The Theory group has four permanent staff members: H.W. Fearing (group leader), B.K.Jennings, J.N. Ng and R.M. Woloshyn. In ad­dition we have research associates and long­term visitors. During this year our research associates have been G. BSlanger, P. Blunden (from August), M. Celio, E.D. Cooper (to August), W. Dickhoff (to July), T. Draper (from October), S. Godfrey (to August), M.J. Iqbal (to September), A. Matsuyama (to March), J. Thaler (to September) and R. Witt- man (from September). Our long-term visitors have been A. Gal (to August) and F. Stancu(from November). Four graduate students, G. Couture, S. Fortin, J. Kopac and R. Workman, are being supervised by Theory group members. Interactions and collaborations are also maintained with theorists from the near­by universities.In addition to their active research program the theorists are involved in a number of laboratory activities including the Long- Range Planning Committee and the Kaon Factory Steering Committee. Organization of the TRIUMF seminars is also also handled by the theory group. This and the summer theoretical visitors program have brought a large number of visiting theorists to TRIUMF this year (listed below).As usual the Theory group has been very active, and we briefly describe in the fol­lowing the specific research proposals under­taken during the year.Scattering and reactions on nucleiPion-nucleus scattering with relativisticnucleons (B.K. Jennings, E.D. Cooper;N. de Takacsy, McGill)The pion-nucleus optical potential is usually derived in the framework of nonrelativistic quantum mechanics. However, it has become popular in recent years to describe nuclear properties in terms of the Dirac equation. It is thus natural to attempt to describe pion- nucleus scattering in this framework. InI. Afnan Flinders A. JacksonR. Barrett Surrey C. JosephK. Bleuler Bonn I. KelsonS. Capstick Toronto L. KliebS.A. Chin Texas A&MF.E. Close Rutherford R. KoniukR. Dalitz Oxford R. KozackJ. Dey Saskatchewan H. LipkinM. Dey Saskatchewan M. LocherM. Dillig Erlangen M. MacfarlaneF. Fiebig Florida Int. A. MaleckiE. Fischbach Purdue 0. MaxwellJ. Fry Liverpool B. McKellarJ. Greben Pretoria H. McManusS. Gurvitz Weizmann G.A. MillerW. Haxton Washington J. MissimerH. He Chalk River N. MobedP. Hoodbhoy Carnegie-Melion M. MoravcsikC. Horowitz MIT S. MoszkowskiW.Y.P. Hwang Indiana H. MutherStony Brook J. Niskanen HelsinkiLausanne K. Olynyk FermilabTel-Aviv R. Perry WashingtonNetherlands A. Polls BarcelonaPlasma Inst. C. Price IndianaYork A. Ramos BarcelonaOhio State E. Redish MarylandWeizmann H. Rolnick BonnSIN M. Samuel OklahomaIndiana StateKrakov J. Smith Stony BrookFlorida Int. J. Speth JulichMelbourne D. Sprung McMasterMichigan State A. Stern OregonWashington M. Theis Frije Univ.SIN AmsterdamToronto I. Towner Chalk RiverOregon E. Truhlik INP, PragueUCLA M. Weyrauch NIKHEFTubingen L.C.P. Wijewardhana Yale70order to do this one must start with a chirally invariant theory. This is particu­larly important if one wishes to study the 1=0 partial wave. A major complication arises in that it is possible to write down many starting Lagrangians which give the same result for physical observables if calculated with sufficient accuracy but which suggest different approximation schemes. In such cases it is easy to mislead oneself. For ex­ample there is much discussion about whether to use pseudoscalar or pseudovector coupling for the pion-nucleon vertex. In fact the answers will be the same in both cases provided one has ensured that the model haschiral symmetry.One of the dominant features of the usual approach to pion-nucleus scattering is the need for an LLEE effect which suppresses the pion multiple scattering. This effect is traditionally ascribed to short-range corre­lations and p exchange. In the presence of an effective mass in the relativistic theory there is an additional suppression of the multiple scattering due to relativistic effects. Thus the LLEE parameter that is put in empirically must be reduced.On the role o f antinucleons in Dirac phenomenology(E.D. Cooper, B.K. Jennings)It has recently been pointed out by Thies [Phys. Lett. 162B, 255 (1985)] that in so far as proton elastic scattering is concerned the iterated spin-orbit potential and the effec­tive mass term give rise to contact terms, in the multiple scattering series, which entail scattering from two nucleons at the same point in space. Experience from pion-nucleus elastic scattering suggests that such terms are spurious and should be set (by hand) to zero. Thies shows that this procedure when applied to the proton elastic scattering gives results which are indistinguishable from those of Dirac phenomenology. Thies then concludes that the antinucleons are forming some sort of short-distance regulator in the theory, and will be irrelevant once some other regulatory mechanism (such as short- range anticorrelations) is used.We have examined [Cooper and Jennings, Nucl. Phys. A458, 717 (1986)] this idea in the con­text of the full Feynman propagator and have shown explicitly that the role of the anti­nucleons is exactly as Thies has said. We find that the antinucleons remove spurious short-ranged structure in the nucleon propa­gator which is connected with relativistickinematics and is of a range of about a fifth of a fermi. Since this 'smoothing at short distances' is inherent in the propagator we conclude that Thies's ideas will not just apply to elastic scattering but to all of hadronic-based nuclear physics. We have quantitatively examined Thies's idea that short-ranged anticorrelations present in nuc­lear matter will provide a regulating mecha­nism but have concluded that they only eliminated about a quarter of the spurious structure.The (p,mr) ground state reaction in a relativisticframework (E.D. Cooper, K. Hicks, B.K. Jennings)Understanding the (p,ir) reaction has been a problem due to the ambiguities in calcula­tions for the bound state neutron, the pion production mechanism and the pion-nucleus final-state interaction. In order to remove two of these ambiguities we have calculated the (p,mr) reaction with the target left in its ground state and the pions having low en­ergy. In the last few years there have been sufficient advances in understanding inter- mediate-energy proton scattering using the Dirac equation that it is possible to calcu­late the distortions for the incident proton and final neutron in a reliable way. Since the pion distortion is well determined at low energies the main uncertainties left in the calculation are due to the production mecha­nism. Thus one hopes that by comparing these calculations with experiment (Expt. 425) one can pin down the reaction mechanism.Pion production in the a  resonance region(E.D. Cooper, A. Matsuyama)A certain set of Feynman diagrams for the process 160(p, Tr+)170(g. s . ) for proton ener­gies between 200 and 450 MeV are summed. This set is collectively known as the stripping (or single nucleon or DWBA) model for the process. The model incorporates all the fea­tures which one expects to be in a model for pion production, and has been quite success­ful in determining angular distribution shapes and their energy dependences on heavi­er nuclei at lower energies. The model has been modified here to incorporate the full A-hole rescattering mechanisms as opposed to previous applications where a local pion potential has been used. The main ingredients of the model, which we feel are essential in any realistic model of pion production, are:1) Realistic proton distortions. These are responsible for the negative analysing power71seen in pion production data at the lower momentum transfers. Pion production is no more or less sensitive to changes in the pro­ton distortions as is proton elastic scatter­ing. Thus a proton optical potential which does not fit the elastic data will not be of any use in pion production.2) A pseudovector production vertex (or equivalent). It is well known that nonrela- tivistic calculations for pion production cannot reproduce the relativistic calcula­tions; therefore, one must directly use the relativistic vertex. This also means that the proton distortions must be calculated relativistically, as must the bound state wave function.3) The pion rescattering mechanism must also be realistic. In particular the pion-nucleus interaction must include a mechanism for sup­pressing high (internal, off-shell) momentum components. In the A-hole model this is done through form factors at the vertices and, perhaps, through employing a finite range in­teracting A propagator. Not just the internal momenta should be suppressed but also the momentum carried by the pion at the first leg of the production process. This is done in a rather cavalier manner by replacing the pion propagator at the first leg of the process by a sum of two propagators which then mocks up the effect of possible p-meson exchange at the first leg.4) The bound state wave function at large momentum transfers must be correct. In the model the Feynman diagram where the proton and pion do not interact is large, i.e. for (p,ir) momentum 'sharing' does not in fact happen. This means that the process is sensi­tive to the high momentum components of the bound state wave function. This wave function is at present taken from Dirac-Hartree calcu­lations and we have no reason to suspect that it is correct out to the momentum transfers of interest in the pion production process.Comparison of the model to preliminary data indicates only qualitative success. The dif­ferences between the model and experiment are greater than can be accounted for by changing the proton distortions; also the energydependence of the analysing power is not cor­rect. Since the model is sensitive to the high Fourier components of the bound state wave function, it may prove possible to modi­fy these so that with one bound state wave function the data at all energies can be reproduced. (The departure of both authorsfrom TRIUMF makes it very unlikely that this last calculation will be done.)The (p,2p) reaction in the Dirac DVZIA(E.D. Cooper, O. Maxwell)The Dirac impulse approximation (including modifications for exchange and Pauli block­ing) using a relativistic 'Love-Franey' type interaction works very well for proton elas­tic scattering. Since the main reaction content of this model is the (p,2p) process one expects it will work just as well forthis process. To check this idea entails doing a full finite range DWIA calculation (relativistic, incorporating exchange and Pauli blocking) for the process. Several new numerical and calculational tools have been developed to do this and at present a comput­er program is being debugged; it is hopedthat calculations will be available by the new year. To optimally test the theory ex­perimental data will be needed in which anangular distribution of two protons is mea­sured where each proton maintains a constant centre-of-mass (three-body) energy.Nuclear structureThe structure o f the deuteron (P. Blunden;P.W. Sitarski, McGill; E.L. Lomon, MIT)We are studying the effects of isobar reso­nances and meson-exchange currents on the electromagnetic structure functions of the d teron. Several potentials with differing isobar content, all of which fit the nucleon- nucleon scattering data up to 800 MeV, are used to construct deuteron wave functions in a coupled channel formalism. With the addi­tion of meson-exchange currents predictions for the form factors and the T2Q tensor polarization are obtained. These predictions will be severely tested by recent experiments at SLAC on the magnetic form factor B(q2) and at Bates on T2Q. There are many potential future applications involving a variety of radiative processes, including electrodisin­tegration, np capture, photodisintegration, etc.Quasielastic nuclear response functions(P. Blunden)Nuclear response functions in the quasielas­tic region provide a detailed picture of nuclear properties complementary to that ob­tained at low energies. Longitudinal and transverse electron scattering response func­tions in the quasielastic region are being72calculated with the inclusion of many-body correlations and meson-exchange currents. As part of a future program the response functions for exclusive processes will also be considered.Another important quantity of interest and study is the ratio of longitudinal and transverse spin response functions seen in proton scattering, which is expected to be enhanced by the pion field.Nucleon form factors in the medium (P. Blunden)The electromagnetic properties of a nucleon in the medium may be very different from those of a free nucleon. If a portion of the nucleon's intrinsic form factor is due to its meson cloud, then this can change considerab­ly in a nuclear environment. Changes in the form factors due to the meson cloud are being calculated for models in which the nucleon is considered a pointlike object as well as various quark potential and bag models.Charge-symmetry breaking forces and the Nolen-Schiffer anomaly (P. Blunden; J. Iqbal, Alberta)Charge-symmetry breaking forces in the np system of the class IV type have now been seen experimentally. Microscopically these forces can arise from an up/down quark mass difference, the electromagnetic interaction of a proton with the magnetic moment of the neutron, p-w mixing, and the proton/neutron mass difference in one-pion exchange. We are investigating the consequences of these forces for the well-known Nolen-Schiffer anomaly, for which they appear to make a sig­nificant contribution.Relativistic nuclear many-body problems(P. Blunden)The relativistic many-body problem continues to be an active area of research in nuclear theory. Previous work has resolved apparent­ly large discrepancies between relativistic and nonrelativistic approaches in the des­cription of electromagnetic observables. Further work on electron scattering studies, both at low energies and in the quasielastic region, is in progress. Weak interactions are still not well understood in these models, partially due to an inadequate treat­ment of the pion and of chiral symmetry. Various chiral Lagrangians are being devel­oped to study weak interactions from a field theoretic standpoint.Nuclear properties near the Fermi surface are not well reproduced due to the inadequacies of the Hartree approximation in this region. The inclusion of higher-order terms in the self-energy, which couple single-particle states to more complicated configurations, is expected to improve the results.Scattering and reactions on nucleonsPion-nucleon scattering: Chiral symmetry andunitarity in the cloudy bag model(E.D. Cooper, B.K. Jennings; A. W. Thomas,Adelaide; P Guichon, Lyon)In order to describe low pion-nucleon scat­tering, especially the s waves, it is neces­sary to develop a model that preserves both chiral symmetry and unitarity. The usual procedure of first calculating a driving term which is then iterated in a Lippmann- Schwinger equation does indeed guarantee uni­tarity but it violates chiral symmetry and in particular the constraints on the scattering lengths. The problem is that although chiral symmetry is preserved order by order in a perturbation expansion, it is not preserved diagram by diagram. The diagrams generated by solving the Lippmann-Schwinger equation are not by themselves invariant under chiral symmetry. On the other hand if we unitarize by equating the driving terms to the K-matrix, we have a procedure that does pre­serve chiral symmetry and simultaneously en­forces unitarity. In addition, by comparing the two approaches we have a check on the importance of higher-order graphs. We find that when we calculate to order (1 /2f)4 the two approaches are in reasonable agreement with each other and with the experimental results.Contribution o f the A(1232) to radiative muoncapture on the proton(H.W. Fearing; D.S. Beder, UBC)Considerable theoretical and experimental ef­fort over a number of years has been directed towards a precise understanding of the so- called 'induced pseudoscalar weak interaction nucleon current' and a determination of its coupling strength gp which is predicted by the Goldberger-Treiman relation. In particu­lar, radiative capture of a muon at rest by either a proton or by a complex nucleus al­lows one to selectively enhance this inter­action since the momentum transfer for near maximum photon energy is much nearer to the73pion pole which dominates gp than is the case for ordinary nonradiative muon capture. There has been renewed interest in this ques­tion in view of new experiments proposed both at SIN and TRIUMF [Expt. 249, G. Azuelos, spokesman] to measure the radiative rate in hydrogen.Precision comparison of theory with experi­ment and extraction of gp is reliable only if one knows that no significant amplitudes have been omitted from the calculations. Previous standard calculations [e.g. Fearing, Phys. Rev. C 21_, 1951 (1980)] include dia­grams with radiation from external legs and from the intermediate pion generating gp but have not included the simplest diagrams involving the A(1232). Thus the aim of this work was to calculate such diagrams to see if they made significant contributions.We have thus calculated as relativistic Feynman diagrams those contributions to radi­ative capture on a proton generated by the permutations of a nucleon-delta-photon vertex and a nucleon-delta-weak vertex. The neces­sary couplings were determined using CVC, PCAC, information on the photon decay of the delta, and constraints obtained by applying the same model to pion photoproduction.We find that these A(1232) contributions change the capture rate near the high-energy end of the photon spectrum by as much as 7-8%. In terms of gp this changes gp by the order of 10%, i.e. by 0.5—0.7 in units of g^. These delta contributions, while not large, are of the order of the expected ex­perimental uncertainty, and thus clearly must be included in precision analyses of the rad­iative capture process. This work [TRI-PP- 86-88] has been submitted for publication.Double scattering contributions to a potential modelcalculation of proton-proton bremsstrahlung(H.W. Fearing, N. Evans)Over the last few years extensive calcula­tions of the proton-proton bremsstrahlung process have been carried out, using both Paris and Bonn potentials [Workman and Fearing, Phys. Rev. C 34 , 780 (1986);Fearing, TRI-PP-86-72]. This has been in parallel with a new TRIUMF experiment measur­ing both cross section and for the first time analysing powers [Kitching et al., Phys. Rev. Lett. 57_, 2363 (1986) and TRI-PP-86-70] The results of the comparison of experiment and theory are really quite striking. For the first time we see clear and unequivocalevidence for effects, presumedly related to the off-shell aspects of the nucleon-nucleon force, which cannot be explained by the pure­ly on-shell soft photon approximation. Furthermore the predictions for the analysing power using both Paris and Bonn potentials agree well with the data. Cross-section data are still to be analysed, but it appears that the potential model calculations give a reasonable description of the process.So far, however, only the single scattering contributions have been included. These come from diagrams involving an off-shell nucleon- nucleon scattering followed or preceded by radiation of a photon. There is anotherpossible contribution, however, the so-called double scattering term arising from an off- shell nucleon-nucleon scattering, followed by radiation of a photon, followed by anotheroff-shell nucleon-nucleon scattering. One can show that the leading term of this con­tribution, in powers of the photon momentum, is proportional to the centre-of-mass momen­tum. Thus by working in the centre of mass, as we do, the term is suppressed. At lower energies the contribution to the cross sec­tion only was calculated by Brown [Phys. Rev.177 , 1498 (1969)] and found to be small.Nevertheless for completeness it should be calculated at the higher energies of interest for the TRIUMF experiment and its effect on both analysing power and cross section determined.Within the context of the existing single scattering calculation the only easy way to evaluate the double scattering term is by straightforward evaluation of the loop inte­grals. Thus we constructed an interpolation routine for the half-shell T-matrix elements, combined the results with propagator and photon vertex pieces and constructed the loop integral necessary. The propagators have singularities which must be extracted and evaluated by hand. When this is done the re­maining three-dimensional integral is regular and can be evaluated by normal quadrature methods.Very preliminary results seem to indicate that these contributions are normally not too large at TRIUMF energies for either the cross section or analysing power. However, there are some specific kinematic situations where the terms are very definitely not negligible. It remains to check these results and their numerical stability very carefully and then evaluate them over the full kinematic range of the TRIUMF experiment.74Constraints on exotic and quark-based potentials and on the off-shell behaviour o f the nucleon- nucleon force (H. W. Fearing)Recent potential model calculations [Workman and Fearing, Phys. Rev. C 34_, 780 (1986)]using both Paris and Bonn potentials seem to agree rather well with new data from TRIUMF [Kitching et al., Phys. Rev. Lett. _57_, 2363 (1986)] for the analysing power in proton- proton bremsstrahlung. Within the context of these calculations one can thus turn the question around and ask what constraints can be put on the off-shell behaviour of the force by these new data. It is also inter­esting to look at some of the new quark-based potentials, some with rather exotic behaviour to see if they can be constrained by the data.Several qualitative features can be obtained from such analysis. First, the results are sensitive to the off-shell behaviour. If one replaces the half-off-shell T-matrices with their on-shell values the results are in drastic disagreement with the data. Second, for the geometries of the TRIUMF experiment results for both cross sections and analysing powers are dominated by the P waves, with the S waves being important only for the cross sections and then only near the forward and backward points. Results from Paris and Bonn potentials are so similar because the off- shell behaviour of the P and higher waves is so similar. If, however, one makes a reason­able change in the off-shell extension of the P waves it reflects in the analysing power results and at some limit can be ruled out by the present data.One can also examine more exotic potentials such as one based on quantum chromodynamics proposed by Kukulin et al. [Phys. Lett. 135B, 20 (1984); Phys. Lett. 165B, 7 (1985)]. It has a strongly attractive central region instead of the usual repulsive core, supports forbidden bound states, and leads to a node in the deuteron wave function. Here it con­tributes only in S wave, however, where its off-shell behaviour is quite different from that of the other potentials. One thus sees little effect in the analysing power results because of the general lack of sensitivity to the S wave. The cross sections at the end­points are increased rather dramatically, however, so that when the cross-section data becomes available it may set limits on the presence of such a potential.Preliminary results of this work have beenincluded in an invited talk at the 11th Int. Conf. on Few Body Systems in Particle and Nuclear Physics, Tokyo [Fearing, TRI-PP-86- 72]. Once the cross-section data from the TRIUMF experiment is analysed, we intend to look more carefully at the Kukulin and other similar potentials to see if quantitative limits can be set on the off-shell behaviour or on the validity of such potentials.Few-body kaon reactions(H.W. Fearing, R.L. Workman)The study of few-body kaon reactions is of particular interest as it allows one to extend knowledge and techniques familiar from the study of other few particle processes to a new realm involving strange particles and at the same time to become familiar with some of the physics which can be done at the pro­posed KAON factory at TRIUMF. During this past year we have concentrated on two partic­ular reactions K“+p Y+y and K~+d -*■ Y+n+y where Y is a A or a E. Both of these, with a A in the final state, are currently being measured at Brookhaven by a group containing a number of TRIUMF/UBC physicists and so one can expect data for comparison. The Kp process is interesting in its own right as a possible source of information about the poorly understood A(1405) and also necessary as input for the calculation of similar pro­cesses on heavier nuclei. The Kd reaction is interesting because it may be possible from it to get information on the A-nucleon force at low energies. Thus one should be able to get the A-n scattering length from it by mea­suring the photon spectrum at high energies just as was done for the n-n scattering length using the process ird ■* nny [Gabioud et al., Nucl. Phys. A420, 496 (1984)].We have now evaluated the Kp + Ay process in a simple pole model using the standard ex­ternal radiation diagrams and other diagrams involving as intermediate states A(1405), I, K*, N*, etc. Our calculation is thus anextension and improvement of the most recent such calculation [Burkhardt et al., Nucl. Phys. A440, 653 (1985)]. Preliminary numeri­cal results differ in detail from those of Burkhardt et al. but agree with the qualita­tive result that the rate is very sensitive to the poorly understood A(1405) resonance. We intend to extend these calculations to examine capture in flight and to investigate the effect of pseudovector instead of pseudo­scalar KN coupling. There have been two other recent calculations of the Kp -*■ Ay process using entirely different approaches.75A cloudy bag model calculation [Zhong et al., Phys. Lett. 17IB, 471 (1986)] uses KN rescat­tering rather than the A(1405) to produce the main effect. On the other hand a nonrelati- vistic quark model calculation [Darewych et al., Phys. Rev. D 32_, 1765 (1985)] gets asimilar rate by neglecting all but the A(1405) and some quark exchange effects. We hope to be able to understand the way these calculations relate to ours and to each other.Some preliminary work has been done also on the reaction Kd Yny, extending and improv­ing the most recent such calculation [Akhiezer et al., Sov. J. Nucl. Phys. 27, 115 (1978)]. Such a calculation requires as in­put the interaction used for the Kp + Yy process. The at rest Kp -*■ Yy reaction is re­quired if we consider the deuteron constitu­ents as frozen, while the full conplexity of the in-flight process is required if we include effects of Fermi motion. Refinements, such as the deuteron D state, remain to be considered.This area of research offers very interesting and exciting possibilities for exploring new physics. Many interesting results are expected in the coming year.Pion-nucleon bremsstrahlung(R. Wittman, H.W. Fearing)With the completion of the SIN measurement of the Tr+p ■+ ir+pY asymmetry parameters at back­ward photon angles and the proposed experi­ment at TRIUMF to do the asymmetry measure­ment at a variety of photon angles, a substantial body of data of a new type will join the existing UCLA cross-section data [Nefkens et al., Phys. Rev. D 28_, 3911(1978)] for pion-nucleon bremsstrahlung. This will provide a direct constraint on the dynamical content of existing theoretical models and allow deeper insight in the devel­opment of new approaches. Current theoretical models have been based on dispersion rela­tion, effective Lagrangian, or nonrelativis- tic rescattering A-isobar approaches, each retaining varying degrees of consistency with the pion-nucleon elastic scattering phase shifts.Although a striking agreement with the exist­ing UCLA data is obtained in the dispersion relation approach of Landi and Matera [Nuovo Cim. 64A, 332 (1981)], the role of theA(1232) is hopelessly buried in integrals containing the P^j phase-shift information.On the other hand, Heller, Kumano, Martinez and Moniz [MIT preprint CTP 1387] treat the distinct contribution of the A very seriously in their A-isobar model and find a reasonable agreement with the data using a A++ to proton magnetic moment ratio between 2.5 and 3.5. Here they admit that the definition of the magnetic moment of a decaying particle may be somewhat model dependent, but their approach still suggests that the bremsstrahlung data can provide a possible constraint on hadronic models concerned with the nature of the internal structure of the A. For example, the SU(6) quark model predicts that the A magnetic moment is given by the product of the proton magnetic moment and the A charge (eA = 2,1,0,-1). Of course, the highermoments (i.e. E2, M3) are zero in this simple model, but can be generated in more realistic models by introducing a deformation mechanism (e.g. D-state component) to the hadronic description.Since these realistic hadronic models (normally providing wave functions in terms of quark constituents) make predictions of the electromagnetic structure involved, it is clearly an advantage to interpret the brems­strahlung data within a model that can make a direct connection with hadronic properties. With this in mind we feel that a re-evalua­tion of the effective Lagrangian approach is in order because of the ease and clarity with which one can introduce the relevant dynamics in this approach. Although such approaches have been unsuccessful in the past, our hope is that improvements can be made in the form of additional dynamics. It also may be important to enforce in a more consistent way the constraints of the two-channel unitarity known as Watson's theorem [Phys. Rev. 95_, 228 (1954)]. This implies that the partial waves of the ttN * irN and irN + irNy reactions having the same initial state quantum numbers carry the same phase. Use of Watson's theorem here assumes that terms in the unitarity equations involving three- (or more) body unitarity may be neglected. Although effective Lagrangian and rescattering approaches have tried to remain true to the Low soft photon theorem [Phys. Rev. 110, 974 (1958)] and thus retain an element of consistency with ir-N scatter­ing, we believe the role of this additional constraint of the Watson theorem is yet to be shown. Therefore, our interest in the sub­ject includes the radiative properties of the A(1232) and vector mesons, as well as re­examining the possibility of using a dynami­cally 'more complete' effective Lagrangian formalism to provide a satisfactory tool for76investigating the tt-N bremsstrahlung amplitude.Our starting point in the construction of such an amplitude is the chiral Lagrangian approach of Peccei [Phys. Rev. 176, 1812(1968)] with a 'mild' adjustment of couplings to approximately reproduce the background phase shifts in the scattering length approx­imation. The resonance width is then intro­duced consistently with the optical theorem (and as not to cause problems subsequently with gauge invariance) to reproduce the reso­nance phase shift. One then appeals to mini­mal coupling to construct an overall gauge- invariant bremsstrahlung amplitude containing 'no free parameters'. At this point other gauge-invariant interaction terms can be added to represent the radiative decay of the A and vector mesons and 'anomalous' multipole moments of the interacting hadrons [i.e. A(1232)] involved.It is our hope that this approach to the ir-N bremsstrahlung problem can: 1) account forthe important physical considerations of the A-isobar model (such as consistency with the ir-N scattering phase shifts), 2) allow the introduction of gauge invariance, relativis­tic effects and additional interactions in a more natural and consistent manner, and 3) provide a straightforward formalism to in­terpret the body of data which will be soon at hand in terms of strong and radiative couplings, thus yielding a valuable overlap with the predictions of these couplings obtained from hadronic models.Precise test for the unitarized pion photoproduction  impulse amplitude in exclusive nuclear reactions (R. Wittman, N.C. Mukhopadhyay)An impulse amplitude for pion photoproduc­tion, suitable for the nuclear problem, is constructed with improved background dynamics and unitarity. The superior features of the amplitude are demonstrated at energies between threshold (Ey = 150 MeV) and theA(1232) resonance energy (Ey = 320 MeV) for the 11+N(y, tt+) ^ Cg.g . transition [Wittman and Mukhopadhyay, Phys. Rev. Lett. 57, 1113(1986)]. The nuclear structure of the A=14 system helps enhance the effects considered. At or near resonance very sizable (20-30%) effects arise both from the unitarity and from the refined treatment of the pionic final state interaction generated in the A-hole approach [Karaoglu, MIT thesis, Sept. 1982 (unpublished)]. Although we find that a number of the discrepancies between theoryand cross-section data [Cottman et al., Phys. Rev. Lett. _55_, 684 (1985); Teng et al., sub­mitted to Phys. Lett. B] are resolved, it appears that in some cases there is still disagreement even when the nuclear structure should be under control [Yamazaki et al., Phys. Rev. C 34_, 1123 (1986)]. We plan to continue our investigation by considering the role of a possible two-step mechanism of coherent ir° photoproduction and subsequent charge exchange. Such a mechanism should be strongly energy dependent, and is under study for a variety of nuclear transitions.Charge-symmetry breaking in neutron-proton scattering (M.J. Iqbal, J. Thaler, R.M. Woloshyn)We have developed a new way of calculating the effect of the neutron-proton mass diffe­rence in neutron-proton scattering. Unlike previous models which rely on the construc­tion of a charge-symmetry breaking potential, we use a relativistic formulation which keeps the mass effects explicitly in the external wave function and internal propagators. The charge-symmetry breaking amplitude is then calculated using the charge-conserving ampli­tude as input. To leading order in charge-symmetry breaking, the charge-symmetry breaking amplitude has two terms, a term which is determined entirely from the on- shell charge-symmetry conserving phase shifts and a term (analogous to a box diagram) which requires an off-shell extrapolation of the charge-conserving amplitude for its evalua­tion. The first term contains effects of the n-p mass difference only in external legs while the latter term also includes the lead­ing effect of differences in mass of internal propagators.The model is being applied to the recent TRIUMF charge-symmetry breaking data. It is hoped that final results will be available soon.Baryon structureOn the nature o f the \(1405) (B.K. Jennings)The A(1405) stands out in quark models as being particularly hard to fit. For example in the most recent constituent quark model [Capstick and Isgur, Phys. Rev. D 34, 2809 (1986)] its energy is off by 200 MeV even after the parameters have been adjusted to fit the other odd parity states. The CBM, which is quite successful in predicting low- energy kaon-nucleon and antikaon-nucleon77scattering, implies very strongly that the A (1405) is a TT-N bound state and thus should not appear as state in the quark models. It has been notoriously hard to get a clean signature of the nature of this state. However, if one considers not only the A(1405) but also other states with the same quantum number the situation becomes much clearer. With the usual interpretation of the states the constituent quark model predicts a large coupling of the A(1800) to the o-ir channel; this is in clear disagree­ment with the analysis of Gopal et al. [Nucl. Phys. B119, 362 (1986)] who have this coup­ling almost zero. On the other hand, if the lowest three-quark state is identified witl the A(1670) and the second three-quark state with the A(1800) the decay predictions become much more in agreement with the experiment. Thus we believe that the A(1405) is definite­ly a K-N bound state.Strong decay o f nonstrange baryons (FI. Stancu, R. Sartor, P. Stassart, Liege)In a recent work [Sartor and Stancu, Phys. Rev. D 31^ , 128 (1985)] we have calculated the positive and negative parity spectrum of the N and A resonances by using a QCD-inspired, semirelativistic quark model. In this model the kinetic energy has a relativistic form and the unperturbed part of the Hamiltonian contains two- and three-body terms derived from a flux—tube quark model. The perturba­tion is the hyperfine interaction with a contact and a tensor term. The only param­eters in the model are the quark mass m and its size A. They can be adjusted such as to give for the N—A splitting and the lowest negative parity state energy values within the experimental error. The Roper resonance appears too high by 100—150 MeV for all sets of parameter values. The rest of the spec­trum is fairly well reproduced.A further test is to analyse the quality of the configuration mixings obtained through the diagonalization of the hyperfine inter­action.We have completed the calculation of the photodecays [Sartor and Stancu, Phys. Rev. D 33, 727 (1986)] with satisfactory resultsboth for the sign and magnitude of the helicity amplitudes.Calculations of the strong decays are less straightforward. There are several models describing the meson-quark coupling. For simplicity, in a first stage, we have chosena pseudoscalar coupling with a recoil term. This kind of coupling has two free parameters and assumes the pion a point-like particle. The resulting pion decay widths have the cor­rect order of magnitude in most cases [Sartor and Stancu, Phys. Rev. D 34, 3405 (1986). As a second stage we consider the pion as having a quark-antiquark structure. Its wave function can be obtained from the same flux- tube model as that of the baryon. We plan to study the pion decay of resonances through the pair-creation model which involves knowledge of the resonance, the pion and the nucleon ground state wave functions.Hadron structure/quark modelP wave meson with one heavy quark (S. Godfrey; R. Kokoski, Toronto)Quark models have achieved considerable suc­cess in describing hadron properties. It is also becoming increasingly clear that the quark model is a reasonable approximation of QCD in the hadron sector. Because of this the effective interquark potential has become a middle ground which lattice gauge theorists can calculate from first principles and quark model phenomenologists can extract from ex­perimental data. Because of this it is use­ful and important to test quark potential models in new situations to further under­stand the interquark potential which can be compared to lattice QCD results. One such laboratory is the study of mesons which contain one heavy quark such as the charmed and beauty mesons. The recent discovery of an excited charmed meson by the ARGUS collab­oration and the prospect of observing more such states adds to the timeliness of such studies.With this motivation we have studied the pre­dictions of the relativized quark model to use as a guide in interpreting experimental data and to discuss the physics which can be learned from the study of these states. Our main conclusions were that: (1) relativisticeffects are important in understanding the properties of P-wave meson, (2) determining the level ordering of these states is impor­tant in understanding the effective inter­quark potential, and (3) the decay proper­ties, especially those of the j=l states, will give further information about the sign and strength of the spin-orbit piece of the Hamiltonian which in turn will give informa­tion on the range of the Lorentz-vector piece of the potential. As such, the study of L=1 mesons will add to our understanding of the78Fig. 87. The reduced vector meson transition strength versus vector meson mass (•)• Also shown are the transition strengths obtained from experimental decay widths (A) and from the constituent quark model (x) of Godfrey and Isgur [Phys. Rev. D 32_, 189 (1985)].nature of confinement [Godfrey and Kokoski, TRIUMF preprint TRI-PP-86-51].Lattice gauge calculationsLattice quantum chromodynamics(R.M. Woloshyn)The amplitude for radiative decay of vector mesons was calculated in quenched lattice QCD. The calculation was done with SU(2) colour and used the Wilson scheme for lattice fermions. The techniques used were similar to those developed for our previous study of the pion electric form factor. The results for the vector meson radiative transition strength is shown in Fig. 77.Lattice calculation o f weak matrix elements(T. Draper; C. Bernard, A. Soni, UCLA)An outstanding problem in low-energy hadronic physics has been the calculation of nonlep- tonic weak matrix elements. Prominent exam­ples include the AI=l/2 rule and the CP violation parameters £ and e1 in neutral kaon decays. Lattice Monte Carlo techniques offer a unique opportunity for performing such cal­culations directly from the fundamental theory. The crucial analytical calculation involves relating matrix elements of the four—fermion operators of the effective weak Hamiltonian in the continuum to their lattice counterparts. We have performed a latticeweak coupling perturbative calculation, which was tremendously more complicated than its continuum analog. We included the case where the operators contain fields which can con­tract with each other; this required the cal­culation, on the lattice, of the historically significant 'penguin' Feynman diagram. We find that the lattice penguin operator is identical to that in the continuum, with only its coefficient differing.Numerical computations to this point have been of reduced K ■+■ it matrix elements. At present we are determining how to remove this approximation by measuring physical K ■+ 2ir matrix elements directly.Grand unified modelsAsymmetries at e+e colliders from Ee grandunified theories (G. Belanger, S. Godfrey)Motivated by the resurgence of interest in Eg grand unified theories due to their possible relevance as the low-energy limit of super­string theories, we studied the effects of the extra Z° bosons predicted by the Eg models on asymmetries in e+e” collisions. We found that the deviations from the standard model allowed by low-energy phenomenology can be quite substantial and that the asymmetries at e+e- colliders offer a sensitive probe for new physics. In particular, the measurements at /s = Mz q are a sensitive test of the mix­ing between the Z° and Z' bosons. In the absence of mixing the asymmetry measurements at > Mzo are sensitive to a Z' of a fewhundred GeV.Superstring phenomenology: Neutrino masses andextra Z°(G. Belanger, D. London, J.N. Ng)By imposing a Z2 discrete symmetry called L-parity on the Eg superpotential we obtain L-parity assignments to the components of the 27-plet chiral superfields of Eg. Very light v < 0.1 eV/c2 is obtained without rapid pro­ton decay. The solution also satisfies the constraints from BSov decays and standard cosmology. In the most attractive scenario a vector lepton of mass ~100 GeV is obtained.In schemes where Eg breaks into a rank 5 group such as SU(3)xSU(2)xU(l)xU(l) an extra Z° is obtained. We investigate the phenome­nology of this added Z° by allowing it to mix with the standard Z°. All neutral cur­rent data, W-boson mass measurements and searches at the Spp^ S collider are used to constrain the mixing parameters. We find that79the existence of Z° can affect the standard Z° width in measurable amounts at the Z° factories. We are now investigating these effects in low-energy neutrino scatteringsuch as vee and elastic scatterings. Inlooking for new physics of Z° high-intensi- ty neutrino beams are required in order to be competitive with measurements at Z° peak.Anomalous magnetic moments of the \N boson (G. Couture, J.N. Ng)The anomalous moment (k ) of the W boson is intimately related to the renormalizability of electroweak theories. In the standard model k=1, whereas in nonrenormalizable model such as composite models of the W boson k can have large corrections of order M^/A where A is the composite scale. We calculated the upper and lower bound to k due to the one- loop quantum effects and found that they are no more than 1.5%. We extended it to an Eg model and new effects of vector lepton also contributed no more than 1% to |k |. We can confidently say that for typical renormaliz- able models Ak receives contributions of the order of a few per cent from fermions. Work is in progress to study the effects of charged Higgs on k .Light scalar or pseudoscalar bosons in fi e transition (G. Belanger, J.N. Ng)We have completed the study of the effects of a family changing the 0+ or 0~ boson, denoted by F, on the y-e system. From the Michel spectrum of y decays we constrain the yeF couplings to be ~10-8. This is more stringent than obtained from (g-2) of electron-muon and muonium hyperfine splitting.Majorana neutrinos and e e collisions (G. Belanger, D. London, J.N. Ng)We propose the reactions of e”e“ to two W  bosons as a new way of testing for the exist­ence of heavy Majorana neutrinos. We ^nd that for a high-energy collider with /s = 1 TeV the signal is competitive with neu- trinoless double 6 decay of Ge, and with higher energies it surpasses the latter experiments. Furthermore, we also calculated the production of right-handed W bosons in these collisions. One feature of e~e” pro­duction of W” pairs is that it is free of large standard model background and thus they are relatively clean experiments to test for Majorana neutrino effects.Electroweak interactionsThe phenomenology o f extra Z°’s in EeGUTs in ep collisions (S. Capstick, Oxford; S. Godfrey)Recently there has been a resurgence of interest in Eg grand unified theories due to the possibility that they may be the low- energy limit of superstring theories. Because of this there has been interest in the phys­ics of extra neutral gauge bosons which is one of the phenomenological predictions of Eg. With this widespread interest in extra Z s it is interesting to compare how well different experiments can probe the parameter space of the model. Motivated by this we have explored how well ep colliders can measure the effects of extra Z°'s and com­pared the results to existing neutral current data and e+e” colliders such as TRISTAN, SLC and LEP. We found that in general HERA will be able to detect extra Z's of much higher mass than e+e~ colliders. If e+e” colliders detect deviations from the standard model, HERA measurements will be important in de­termining the parameters of the new Z°. Crucial to these results was the need for high statistics, and we concluded that it will be more important to enhance statistics rather than increase the energy in looking for effects due to extra Z's.AstrophysicsNeutrino emission processes in hyperon-populated neutron stars (O. V. Maxwell, Florida International)Several neutrino emission processes that can occur in neutron stars containing hyperons, nucleons and electrons are examined in low­est order in the strong and weak interac­tions. These processes are of two types: neutrino pair bremsstrahlung processes of the form AB ->• CDvu, where A,B,C,D denote baryons, and URCA processes of the form AB + ACev. The processes are assumed to occur in a dense, degenerate environment subject to charge and beta equilibrium in which the electrons are extremely relativistic and the baryons non- relativistic. To obtain expressions for the emissivities an independent particle des­cription is employed in which the baryon Fermi energies and momenta are related through the introduction of effective masses, short-range correlations between the baryons are ignored, and the strong baryon-baryon interaction is approximated by a one-pion- exchange potential with coupling coefficients determined by symmetry considerations. Numerical results are then extracted for a80number of neutron star scenarios character­ized by fixed values of the fractional baryon populations and the electron-to-baryon ratio. In all scenarios considered it is found that the URCA processes dominate the bremsstrah­lung processes and that among the URCA pro­cesses the most important processes are those involving nucleons alone, provided the nuc­leons in the interior are normal. If either the neutrons or protons are superfluid, how­ever, then processes involving hyperons will contribute the largest emissivities, to such an extent that in stars with superfluid nuc­leons the presence of hyperons will signifi­cantly accelerate the cooling. Possible improvements in the calculations and their consequences are briefly discussed in the conclusions.Muon spin rotationNew m ethod to calculate the muon polarizationfunction (M. Celio)For the last few years muon-spin-relaxation processes in zero and low field have been the focus of much theoretical and experimental work in the field of ySR. In particular, zero-field measurements have been performed to investigate the diffusion and trapping of positive muons as a function of temperature in various metals. Theoretically these pro­cesses are difficult to treat because the internal dynamics of the nuclear spin system interacting with the muon has to be taken into account and, in general, no simple per­turbation expansions are possible.In order to solve this problem we developed a new technique [Celio, Phys. Rev. Lett. 56, 2720 (1986)], which is based on the Trotter formula and leads to exact results even when the dimensionality of the composite spin sys­tem is very large. This method has already been applied to the analysis of ySR experi­ments in copper as well as in single crystals of NaF, LiF and CaF2. The introduction of this technique will greatly facilitate the analysis of zero-field, low-field and yLCR (muon level-crossing resonance) results interms of fundamental microscopic parameters instead of phenomenological quantities. This approach might be generally applicable to describe different interstitial or substitu­tional particles in solids.Theoretical analysis o f muon level-crossingresonance (iiLCR) experiments (M. Celio)The possibility of using level-crossing reso­nance (LCR) spectroscopy in ySR was first pointed out by Abragam, and the first suc­cessful yLCR experiment was performed in copper [Kreitzman et al., Phys. Rev. Lett. 56, 181 (1986)]. Recently it has been proved That a similar technique can allow precise measurements of nuclear hyperfine parameters in muonium-like systems. This was first shown for the case of a muonated free radical • C6F6-Mu [Kief1 et al., Phys. Rev. A 34, 681 (1986)], and very recently this spectroscopy was applied to muonium defect centres in semiconductors.Several numerical calculations have been per­formed in order to calculate the explicit time dependence of the muon polarization for cases where the muonium atom is interacting with neighbouring nuclear spins. These cal­culations allow one to extract the nuclear hyperfine parameters from the positions of the measured LC resonances. Moreover, they can be useful to check approximations which may be relevant in the analysis of a particu­lar experiment.Typically an integral yLCR experiment is complemented by time differential measure­ments, which also contain a considerable amount of information. Some of our previous calculations based on a master equation were generalized to describe the dynamics of muonium interacting with nuclear spins at a LCR. Within this approach it is possible to describe a number of depolarization mechan­isms, such as electron-, nuclear-, or muon-T1 relaxation processes, or partial averaging of the anisotropic part of the nuclear hyperfine interactions due to the diffusion or tumbling of muonium.81APPLIED PROGRAMS DIVISIONINTRODUCTIONThe Applied Programs continue to mature and develop along the lines of cancer treatment with pions, the production and supply of radiopharmaceuticals, and the use of these pharmaceuticals for positron emission tomog­raphy. The pion therapy program has in­creased the dose per fraction very close to the normal tissue tolerance dose. In some cases with this high dose the tumour re­gressed to the point of not being detectable on CT scans. There has been continuing growth and development at the CP-42 cyclotron where most development has been directed towards increasing the operational reliabili­ty of the facility, thus increasing the out­put and maintaining a reasonable operator dose level. The radioisotope sales program over the year was profitable and AECL now offers radiopharmaceuticals made from TRIUMF- produced radioisotopes. Radiopharmaceutical kits that are very efficient have been developed by the TRIM group as well and are in use in the local area hospitals. Though only using 5% of the CP-42 cyclotron beam time to produce PET radioisotopes the program has established an active clinical program in brain disease study.BIOMEDICAL PROGRAMThe number of cancer patients treated in 1986 was significantly lower than that of previous years because the biomedical channel was out of commission during the major run in July and August. With the relatively small number of patient treatments conducted the treatment regimens were therefore kept essentially the same as the latter half of 1985 for both brain and pelvic patients, in order to accu­mulate better statistics. A list of patient treatments for 1986 is shown in Table XII.The run in January had only 14 days of sched­uled high intensity operation, which was not suitable for our basic 15 daily fraction treatments for both brain and pelvic tumours. However, by special arrangement with Opera­tions four extra treatment days were arranged between Christmas 1985 and New Year 1986, so that four treatments were successfully completed in the January run.In February the 2 MB hard disc drive of the Nova II computer system began to malfunctionintermittently. The previous failure in September 1985 destroyed some important treatment files, but the disc contents were successfully reconstituted into another disc from backup copies stored in floppydiscettes. A new disc drive of 20 MB wasordered as a replacement because it wasdifficult to obtain parts and expertise for the repair of this 13 year old 2 MB discdrive. This incident also reminded us that most of the other biomedical channel control hardware, including the Nova II computer, were over 10 years old, and hence were well beyond the usual half-life of this equipment. A comprehensive hardware upgrading plan was therefore developed and presented to the hospital administration for consideration in the summer.At the end of June, just about 2 weeks before the scheduled July-August run, the first dipole of the biomedical channel was acciden­tally pushed off alignment during repair work on the adjacent M20 channel. This resulted in several severe vacuum leaks at the front end of the channel. A quick repair was attempted but failed. Since a proper repair required some detailed planning and was expected to take several weeks, it was there­fore decided to blank off the M8 channel for the July-August run and to resume repair in September. This was the first time that the biomedical channel missed a complete block of high intensity beam operation.Table XII. Summary of TRIUMF pion patient treatments for 1986Run Patient Fraction finished/intendedTotaldaysDose/fraction (it- rad)Jan 85-32 brain 15/15 18 240Apr- 86-1 brain 15/15 22 220Jun 86-2 brain 15/15 21 22086-3 brain 15/15 29 22086-4 brain 15/15 22 22086-5 pelvis 15/15 22 22086-6 brain 15/15 23 22086-7 brain 15/15 21 20086-8 groin­skin10/10 11 250Nov- 86-9 pelvis 15/15 23 220Dee 86-10 brain 15/15 23 2208 2Repair to the M8 channel was put on high priority and began as soon as the September shutdown commenced. Due to the high back­ground radiation level near the T2 target area the biomedical staff members could not take part in the repair of the M8 channel because they had all received high levels of neutron exposure doses working in the M8 cave during routine patient treatments. The repair was completed on schedule in early November and two more patients were treated before the end of the year.Over the last several years the dose used for the brain and pelvic tumour treatments has been gradually increased to the present level of 33-36 Gy in 15 fractions. One of our goals is to reach normal tissue tolerance doses so that we can deliver the maximum possible tumoricidal doses. For the past year most of the brain tumour patients received a daily dose of 220 cGy for 15 fractions. Preliminary analysis indicates that the median survival for the patient group treated with 240 cGy is not as good as expected from extrapolation from results obtained at lower pion doses. However, the 240 cGy group of patients had a mean age of over 64 years, which is signifi­cantly higher than the previous groups and could account for some of the differences. However, it was also observed that one of the patients treated with 240 cGy had shown evi­dence of cortical-neuronal degeneration dueTime (d ay s )Fig. 88. Results of glioblastoma treatments at TRIUMF since 1982. Solid line: convention­al treatment with X-rays or 60Co. Dot-dash line: pions at 1.7 Gy per fraction, 15 frac­tions. Dotted line: pions at 2.0 to 2.2 Gy per fraction, 15 radiation, which suggested that we might have reached tolerance doses. On the other hand, the tumours of two of the brain patients regressed after pion treatment to the extent that they were no longer detect­able in the CT scans. This was the first time such a marked response had been observed.Figure 88 shows the results obtained for the brain tumour treatments at various pion doses together with the results obtained from con­ventional 60Co irradiation at the A.M. Evans Clinic. The data clearly indicate that there is at least a two-fold increase in median survival rate for patients treated with pions. Furthermore, it is also observed that patients treated with pions generally have good quality of life for the survival period. Many were able to return to work and to their normal lifestyle.42 MeV CYCLOTRONOperation1986 was the third full year of cyclotron operation. As during the previous years the machine has been operating every week of the year. It delivered 729 mAh of beam, an in­crease of 14% over the second year. The machine was used for 95% of the time forradioisotope production of AECL's Radiochemi­cal Company and 5% of the time for the pro­duction of positron emitters for the PET program at the UBC hospital.Figure 89 illustrates the weekly beam produc­tion during the year. At the beginning of the year the production was actually only about 70% of that in 1985 over the same period, and it was difficult to meet thescheduled production, which was based on an average beam current of 140 yA on the solid targets. This was caused by poor H- produc­tion in the ion source. Towards the middle of the year this improved somewhat and during the last quarter normal production was restored and eventually exceeded. The average beam current as used for scheduling was in­creased to 150 yA but in spite of this the actual production exceeded scheduled produc­tion, as is reflected in Fig. 89. Duringthis quarter the cyclotron operating time wasalso increased from 142 to 149 h/week by re­ducing the maintenance period. Both the higher beam currents and longer operating hours are responsible for the high production rate during the last quarter.8325a  :  Actual Dose + :  Scheduled DoseAccumulated Dose Scheduled DosemAh/ week -20 30SCHEDULEFig. 89. 42 MeV beam delivery 1986.!■“ A CCTV system for all beam lines in the cyc­lotron vault and target cave was completed to allow remote viewing of the beam spots as .8 projected on retractable fluorescent screens.An external beam wire scanner system that ,6 uses an inexpensive computer rather than an oscilloscope to display the beam distribution  ^ was completed and the design was published.A 15-foil extractor carousel was developed to replace the present 4-foil carousels. This will eventually allow 2 to 3 weeks of operation between foil changes rather than the present 1 week. The first 15-foil carou­sel was installed on December 26 and worked flawlessly for the first two-week period.The improved target stations, installed in 1985, are giving very satisfactory service and the design has been published.FailuresBy the end of May there had been another six insulator failures, after which the four high voltage insulators were routinely replaced every six weeks. Since then there were only two more insulator failures by year-end.A major repair was undertaken early in July: the changing of the upper pole tip O-ring. This vacuum seal had been leaking intermit­tently for quite some time. The repair required careful scheduling to limit thedowntime to only two days, at the end andbeginning of weeks 28 and 29. The repair wascompletely successful. Whatever it was that made this vacuum seal leak it was not radi­ation damage, which was in accordance with our estimates.With the heavier running schedule radiationdamage induced equipment failure is becoming an appreciable factor affecting downtime. Examples are the solid state rf tube filament rectifiers and plastic cooling tubes and insulation around the target stations.ImprovementsA PDP 11/23 computer was acquired to update CP-42 software and to include the various target cave equipment built in house. It will also serve as back-up for the present PDP 11/03 control computer. An IBM compat­ible personal computer package was acquired to replace the Apple compatible PC to improve recordkeeping and prompt reporting of pro­duction data, radiation exposures, spares inventory, preventive maintenance, etc.RADIOISOTOPE PROCESSING (AECL)The CP-42 cyclotron, which is the workhorse for isotope production, has performed such that the program has been able to become a profitable operation for the first time in its existence.The world's largest production run of 123I _ 466.2 GBq - was produced during the year using the 12l+Xe gas target technology. The 123I material produced is shipped to hospi­tals, universities and radiopharmaceutical manufacturers in Canada, United States, Japan and Australia.The target window for the i23I has received a maximum of 40,000 pAh, which is a record, and appears to still be in good condition. The patented i23I target production technology was sold in Europe and successfully trans­ferred during the year.The i23l labelling program initiated last year saw the regulatory approval for the man­ufacture and commercial sale of 123I sodium iodohippurate in Canada. Production and de­livery of 123I sodium iodohippurate has been regular during the last quarter of the year.Also during the year 123I N-isopropyl-p-iodo- amphetamine was submitted to the regulatory authorities as an investigational new drug (IND). 123I N-isopropyl-p-iodoamphetamine is a new radiopharmaceutical which is used for early detection of cerebral vascular attacks. Work is continuing to further expand the 123I84radiopharmaceutical product line and facilities.AECL negotiated a contract with the Indo­nesian Atomic Energy Agency (BATAN) which requires training of several people in cyclo­tron operation, cyclotron isotope processing, physics and nuclear medicine procedures. During the year 5 individuals have attended the site to receive training on the various aspects of operation.The consistent performance against plan on the 500 MeV facility has allowed AECL to meet commitments for 127Xe. Only a couple of times during the year were back-up arrangements with Brookhaven National Laboratory called upon.500 MeV RADIOISOTOPE PRODUCTIONThe year 1986 was the seventh year of ope­ration for the 500 MeV isotope production facility. The facility performed without failures and received a total of 183 mAh of beam. The use of the facility over the years has been as follows:YearmAh to facilityTargetsirradiated1980 51 401981 56 531982 120 491983 142 701984 172 821985 218 901986 183 106POSITRON EMISSION TOMOGRAPHY (PET)The PET program continued in 1986 with sup­port from the Medical Research Council of Canada via renewal of Special Project grant SP-7. By mid-November a total of 408 human subjects had been scanned in research pro­grams on dementia, movement disorders, epi­lepsy and brain tumours.The UBC/TRIUMF PETT VI tomograph, built in 1981-82, encountered some mechanical diffi­culties this year. As a result, the gantry support system was partly rebuilt to an im­proved design, and this alleviated most of the operating difficulties. The associated computer system was upgraded by the addition of more peripherals, in part in collaboration with the UBC research program on multiplesclerosis. The package of image reconstruc­tion and analysis software was extended, and work continued on software for automatic scanning according to a pre-specified protocol.The PET physics research program continued this year with an investigation of the opti­mum axial sampling density required for the production, from the axial images (produced by the camera), of sagittal and coronal slice images via software manipulation. In addi­tion, studies of the behaviour of the tomo­graph system at the very high counting rates required for measurements with 150-labelled radiopharmaceuticals were initiated, follow­ing the observation of some image distortion under these conditions. Studies led to modi­fication of the camera electronics with sig­nificantly improved high-rate performance.The program of chemical research into synthe­sis of new scanning agents continued this year; the routine synthesis of 18F-labelled 2-deoxy-2-fluoro-D-glucose was being accom­plished throughout the year by computer con­trol, freeing personnel for the research pro­gram and reducing significantly personnel radiation exposures. The routine synthesis of 18F-labelled 6-fluoro-L-dopa was partially automated this year also.The synthesis of 11C-labelled L-dopa was fully worked out this year and the first batches delivered to the PET camera in the UBC Hospital. Also, 68Ga-labelled gallium- EDTA (an agent for measuring disruption of the blood-brain barrier) also became avail­able via a generator system at year's end. Research continued into the production of 75Br for incorporation into bromospiperone, of which the chemical synthesis was also established.Research concerned with the origin of Parkin­son's disease continued this year, with more studies of subjects who had been exposed to the neurotoxin MPTP travelling to UBC from California. Also this year a number of sub­jects were scanned from Guam, where the disease is often associated with dementia and ALS (also believed to be the result of neu­ronal degeneration) and where the incidence of Parkinson's disease is 50 to 100 times greater per capita than in North America. The subjects travelled to Vancouver with the as­sistance of the US Navy and included both Guamanians with the disease and others who had been exposed to the same environment but were asymptomatic. Figure 90 illustrates the856-fluorodopa uptake deficit observed in a subject suffering from the disease, in com­parison with a subject who had been exposed to MPTP, a patient suffering from idiopathic Parkinson's disease and a normal volunteer.In the research program on Alzheimer's disease we have begun this year to be able to correlate reductions in regional cerebral glucose metabolic rate in subjects suffering from the condition with the corresponding neuropathological changes observed in such subjects postmortem. An excellent correla­tion was found to exist, and also between PET glucose scan data and the severity of the disease symptoms.In the research program on Huntington's disease emphasis has shifted from the use of PET in symptomatic patients to its use in presymptomatic diagnosis in at-risk individu­als. The PET studies have been performed in conjunction with DMA studies which are about 95% accurate at identifying carriers of Huntington's gene in suitable families. The results indicate that decreased caudate metabolism does indeed precede the develop­ment of clinical symptoms.In research on epilepsy patients with focal epilepsy have been studied, before and after surgical treatment by corpus callosum sec-UBC/TRIUMF PET PROGRAMMPTP G U A M A N IA NFig. 90. PET scans conducted with L-6-18F- fluorodopa. Upper left: a normal subject.Upper right: a subject suffering from idio­pathic Parkinson's disease. Lower left: an asymptomatic subject who had ingested MPTP neurotoxin. Lower right: A Guamanian subject suffering from the Guam dementia-Parkinsonism complex.tion. In addition, the possible relationship between antiepileptic medication and cerebel­lar metabolism is under investigation. The epileptic focus is often evident as a hypo- metabolic area on inter-ictal scans.Publications from this program which appeared in the literature or which were accepted for publication this year are listed in Appendix A.TRIM PROGRAMSignificant activities for TRIM this year were the development of a series of radio­pharmaceutical kits for the 123I and the suc­cessful completion of a fast neutron facility for biophysical studies on beam line 2C.During 1986 TRIM has made 123I-labelled radiopharmaceuticals continuously available to Vancouver hospitals through the introduc­tion of kits developed at TRIUMF. The kits enable rapid and quantitative incorporation of 123I and are believed to be the most eco­nomical means of obtaining these radiopharma­ceuticals. Thus a unique opportunity has been available to hospitals close to TRIUMF. Hippuran, MIBG and IMP are presently avail­able and additional pharmaceutical kits are in the development stage.Heart research at the local clinics con­tinues, utilizing 123l fatty acids from TRIM as one measure of heart physiology. A Canadian patent has been awarded for a pro­cess which allows useful heart imaging, fatty acid analogues to be obtained. The clinical research is being pursued in two directions. One study compares the actual physical state of the heart with typical patient scans. The other studies seek to improve techniques of heart surgery, in particular cardiaplegia, and to provide better care of the donor heart during transplant procedures.In a different direction, members of the TRIM group have completed a fast neutron biomedi­cal facility on beam line 2C capable of delivering 0.05 Gy per microampere of 100 MeV protons. The mean energy of these neutrons is substantially higher than has been previ­ously used for radiotherapy at other centres. An MRC grant has been obtained by the CCABC- TRIUMF Biomedical group to characterize this facility in terms of its biological effec­tiveness. In vitro and microdosimetric studies are scheduled to begin in the summer of 1987.86CYCLOTRON DIVISIONThe reliability of operations continued at levels comparable to those of 1984 and 1985 with 4925 h of delivered beam over 5549 h scheduled, corresponding to a reliability factor of 88.3% (89.5% in 1985 and 88.5% in 1984). A total beam charge of 302 mAh was delivered, ~10% below the 332 mAh record of1985. 23.5 weeks were dedicated to highintensity beam production (25.5 in 1985),14.5 weeks to polarized beam (13.5 in 1985), 14 to shutdowns (13 in 1985). Substantial improvement programs, modifications and beam tests were carried out during the shutdowns, both in the cyclotron and along the vault beam lines. The most significant were the installation of eight new resonator segments in the cyclotron, the reconstruction of beam line 4 for longitudinal polarization and the demonstration of 85% transmission efficiency through an electrostatic deflector at 450 MeV. Tests and installations proceeded on schedule, without any major disruption to beam delivery. Problems leading to a reduc­tion in beam charge and hours were related to the maintenance of systems, such as inflector and vacuum, which in previous years had beenfairly reliable. The traditional histogram, showing the beam charge delivered per month and the beam charge and hours delivered per year since shortly after the start-up of the facility, is shown in Fig. 91.The emphasis was again on developments and improvements. Since a satisfactory level of reliability was achieved and maintained over the last four or five years, it becomes important now to rapidly develop cyclotron beam capabilities not only toward making the machine a suitable injector for a possible post-accelerator, but also toward higher beam intensity, better quality, stability and new types of beams, in order to keep TRIUMF at the forefront of research in its friendly competition with the sister facilities in Europe and in the U.S.Two major projects were completed in 1986. Firstly the feasibility of an optically pumped polarized source capable of providing at least a 5 pA, 60% polarized beam extracted from the cyclotron was definitely proven in the laboratory. At the same time the thirdI H  I I I M I I I I I I ’ I 1 I I I I ■ I I I I I I I I M I I I I M  n  I I I I I I I M M I I I M I M 11 I I 11 I I I1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986Fig. 91. Beam charge delivered and hours of operation over the past eleven years. Milestones in extracted peak current are also indicated. The histogram shows the charge delivered per month.87ion source was completed. The realization of(1) a new large 300 kV high voltage terminal,(2) a new cw high brightness multi-cusp ion source with 2 iA H" current within an accep­tance of 0 .2 t t  mm-mrad, and (3) a new portion of injection line connecting the new terminal to the existing injection line system, were the major accomplishments of this latter project. During the summer a 1 mA equivalent H- beam was transported with 80% efficiency to the spiral inflector at the entrance of the cyclotron. By year-end the system was reliable enough to allow using the third ion source for beam production. Beam studies with the third source and terminal continued both for cw and pulsed beam conditions. Currents around 150 pA cw and around 270 pA pulsed were confirmed.The encouraging results obtained from the optically pumped polarized source prompted the creation of a two-year major project aimed at the installation and commissioning of a new operational polarized source in the third terminal. A fourth section of injec­tion line will connect this source to theexisting injection line. The third terminalwill therefore initially be equipped withboth the optically pumped polarized ion source and the high brightness H- cusp source. Once the polarized beam will have been injected and extracted from the machine (it is hoped within fiscal year 1987-1988) and the 5 pA, 60% polarization expectations will have been confirmed, it is planned toincrease by an order of magnitude the polari­zation (to 80%) and intensity. This will require a more intense magnetic field (2.5 T instead of 1.2 T) in the region of the sodium vapour and more laser power.The prospect of high intensity beams from the new sources led by year-end to a proposal to study the feasibility of simultaneous accele­ration and extraction of both polarized and unpolarized beams. A design study was initi­ated. Recently a 500 pA upgrade task force was created to co-ordinate, oversee and effectuate tests and improvements which will allow the higher intensity available at injection from the new cusp source to be actually accelerated and extracted for beam production. This task force will work in close contact with the 1AT2 upgrade task force which has the target area as its major responsibility. Extraction of a 300 pA cw beam from the cyclotron is planned for 1988.The feasibility of extracting an H- beam from the cyclotron was clearly demonstrated when85% of the accelerated H~ beam could be de­flected, at 450 MeV, to a radial position about 1 cm outside the internal orbits, by using an electrostatic deflector (DCD) and a radial 11.5 MHz rf deflector in the vr=1.5 region. The 15% fraction which could not be deflected by the DCD was prestripped with an upstream foil and extracted through a differ­ent path without causing induced cyclotron activation. Several aspects of the H~ extrac­tion process still need to be finalized, such as the design and construction of a magnetic channel with a 1 to 2 cm narrow septum, the efficient deflection of a beam with 100 pA current, and the capability of the DCD to hold the necessary voltage in presence of a high intensity beam. Work is proceeding with tests and prototyping toward the practical optimization of these goals. Also, design, prototyping and tests are proceeding for the 92 MHz booster cavity and amplifier which will double the energy gain per turn, increase the H” extraction efficiency and reduce the electromagnetic stripping losses.Perhaps the most significant 1986 highlight for the Cyclotron Division was an important breakthrough on the front of the resonator improvement program. This program had origi­nally been set up to eliminate the frequent failures related mainly to excessive tempera­tures generated by rf leakage on the norr- cooled beam side of the resonator segments. The replacement of all eighty segments with more rigid, beam-side and rf-side cooled units had been envisaged. During 1986 improved understanding of the coupling mech­anism between rf cavity and beam cavity was achieved. By year-end after the addition of (1) eight new, more rigid, stable segments in mechanically and electrically strategic posi­tions, (2) remote control of ground arm tips, and (3) on-line displays of leakage voltage and temperatures, it was possible to reduce the leakage voltage by a factor of 10, to values almost everywhere below the multi- pactoring threshold of ~300 V. This resulted in significantly reduced temperatures and, hence, in greatly improved mechanical stabil­ity. The complete replacement of all remain­ing 72 old resonator segments is no longer a compelling requirement.One or more sets of eight segments will have to be replaced with new ones in order to improve the stability, low leakage and tuning flexibility required for third harmonic flat- topping. Most important for third harmonic operation will be a new set of central region segments permitting an acceptable level of88voltage uniformity along the dee gap and the reduction of cross currents. Model studies and prototyping of the new central region are under way. Installation could begin in two years. In the meantime flat-topping tests are being carried out in a pulsed mode in the cyclotron. Tuning and control methods developed in a full-voltage teststand in the laboratory are being implemented in the cyclotron. A flat-topped waveform in the pulsed mode has been demonstrated at 60 kV.The Cyclotron Development group was able to capitalize on the lower rf leakage, by reac­tivating a set of non-intercepting internal beam phase probes which had previously failed to give reliable information. In addition to lower rf leakage, special shielding and filt­ering techniques contributed towards achiev­ing a ~1° resolution at ~1 pA beam current levels. The Probes group installed a new ac­ceptance-defining flag in the central region and a special carousel on the beam line 4 ex­traction probe. This permits selection of the appropriate extraction foil type without foil disposals or hands-on cartridge exchanges.The Vacuum group also welcomed the lower rf leakage, since in the past interference with cryopanels had often caused an increase in the cyclotron residual pressure. With four new cryopumps (16,000 £/s in total for Hg) substituting for four old diffusion pumps (4000 a / s) the cyclotron can now operate without liquid nitrogen traps and without danger of diffusion oil contamination. Ope­rational pressures in the low 10~8 Torr region can now be achieved one or two days after letting the tank to air and can be maintained thereafter more regularly.To conclude, good progress and several inter­esting results highlighted this past year. In cases where systems failures, insufficient budgets or high exposure regions caused new unexpected and difficult circumstances, the groups reacted promptly with ingenuity so that beam production could still be maintained at high levels. Developments on ion sources, rf and H- extraction are pro­ceeding in order to upgrade beam performance and beam capabilities. However, upgrading of deteriorating or obsolete systems like the B20 cryogenerator, the vertical injectionline, the central control system, probes and remote handling equipment will requiremore attention in order to augment the ope­rations efficiency and to maintain or improve present levels of machine reliability and utilization for the future.BEAM PRODUCTIONBeam production in 1986 was 302 mAh, which was down slightly from 1984 and 1985. There are basically two reasons for this. The total number of hours of operation, shown in Fig. 92, was reduced from 5407 h in 1985 to 4925 h in 1986, and beam was run at reduced currents because of the inflector difficul­ties in the spring. The unpolarized beam production is shown in Fig. 93.The number of hours of polarized operationcontinued to rise from 1647 h in 1985 to 1894 h in 1986, which is approximately 14 weeks. There was very little downtime that affected polarized operation.0)C LoWEEKFig. 92. 1986 hours of operation.8 00 070006000500 04000300020 C 010000c lo01W E E KFig. 93. 1986 beam delivery.2001751501251007550250H  P o larized Q  U npolarized5548 Hrs. Scheduled 4925 Hrs. Delivered89Total Beam Charge (pA Hrs.)The operating hours for 1986 are broken down in Fig. 94. The total downtime has dropped slightly from 500 h for the last two years to 457 h for 1986, but different systems have failed. The major downtime this year was due to the inflector sparking following the spring shutdown, a mechanical failure of the B20 cryogenerator in August, and numerous water leaks on magnets around 1AT2 target throughout the first half of the year. The total length of shutdowns also increased by approximately two weeks.Table XIII shows the total beam scheduled and delivered to each experiment during the year.A brief summary of the year's operation is:Cyclotron ONBeam to experimentersDevelopmentTuningCyclotron OFFShutdownMaintenanceDowntimeOverhead, start-up, etc.4355 h 460 1104925 h2595 h 474 457 2853811 hCYCLOTRONCyclotron developmentIn 1986 major advances were made in under­standing the mechanism and successfully re­ducing the leakage of rf power into the beam gap. This improvement of the leakage field in the beam gap and in the whole cyclotron is having a positive effect on the overall per­formance and reliability of operation. In addition to the improved temperature distri­bution on the resonator strongbacks, all the diagnostic and control devices operate with much reduced rf pickup. Measurements of the internal beam phase are now possible. The main cavity has been tuned simultaneously for the fundamental and third harmonic resonances and a flat-topped waveform was achieved in the cavity in a pulsed mode. Enhancements in computer data acquisition and analysis facil­ities were instrumental in carrying out these developments.STARTUP OVERHEAD OTHER MAINTENANCE (47 4 )h o u r s  OPERATION (47;SHUTDOWN (2595)(45 7 )BEAMOPERATION(49 25 )MAGNETSSERVICESCONTROLS11) DIAGNOSTICS 10) TRIPS ■" OTHER SAFETYCORRECTION PLATES TARGETS TNFINFLECTOR(184) VACUUMFig. 94. Operational hours for 1986.due mainly to two parasitic modes excited in the cavity formed by the vacuum tank and the region between the resonator strongbacks. A frequency scan of the modes near the driven frequency is shown in Fig. 95, with the spatial distributions for the TM31Q and the19 21 23Frequency (MHz)rf studiesTheoretical, model and machine studies have shown that rf leakage into the beam gap isFig. 95. Frequency scan of resonant modes in the beam gap near the operating frequency and spatial distributions of the TM310 and TM^10 modes.90147241248249249266111273276281285287288297300300301302304311311319323324325327330332337338339340340341342344346347349349350351352352352354354355357358359361Table XIII. Beam to experiments for 1986.Scheduled Deliverednannel h h (pol) pAh h h (poi;M20 128.0 6498.0 57.7M20 144.0 20160.0 159.6Ml 3 1630.0 194160.0 1377.7M9 2014.0 247678.0 1702.0Ml 3 46.0 6440.0 45.54B 92.0 124.04B 116.0 92.3M20 57.5 8050.0 37.8M20 79.0 11060.0 79.8Mil 74.0 10360.0 85.4Ml 3 498.0 58298.0 375.0IB 179.5 103.4Ml 5 58.0 8120.0 54.3M20 345.0 48300.0 344.74A 9.64B 35.0 150.0 31.2 86.8IB 438.0 321.94B 173.0 155.2Ml 5 501.0 47280.0 326.04A 12.0 2.64B 12.04B 68.5 58.7Ml 3 127.0 17780.0 125.14B 323.0 206.7M20 404.0 43060.0 228.2Mil 809.0 90400.0 545.64B 46.0 58.74C 227.5 181.7Mil 1278.0 156318.0 1162.44B 23.0 16.4Ml 5 371.5 52010.0 344.9M15 150.0 21000.0 157.9M20 196.0 27440.0 156.84A 151.0 134.2Ml 5 197.0 27580.0 187.94B 127.0 136.7Ml 5 139.0 19460.0 132.7M15 69.0 9660.0 69.4M15 69.0 9660.0 65.9M20 130.5 18270.0 110.7Ml 3 220.0 30800.0 212.3Mil 277.0 38780.0 278.14C 257.0 171.3 3.04A 23.0 31.44b 115.0 108.54b 46.04B 138.0 116.34B 23.0 58.0 74.54B 53.0 68.4M15 214.0 18780.0 225.34B 254.0 225.3M15 46.0 6440.0 46.491Table XIII (cont'd)Scheduled DeliveredExperiment* Channel h h (pol) pAh h h (pol) pAh361 M20 57.0 7980.0 57.3 6483.3362 M20 138.0 13110.0 111.7 10982.4364 M20 474.0 66360.0 450.3 57489.1365/249 M9 898.0 114540.0 819.1 82776.4366 4B 127.0 103.5367 M15 751.0 93718.0 611.2 74219.2368 4B 69.0 65.1371 M20 405.0 53550.0 378.0 43708.1373 Ml 3 127.0 17780.0 132.4 16949.7374 M13 139.0 19460.0 133.1 18148.1375 Mil 254.0 35560.0 237.9 29723.0376 4B 289.0 218.9377 Mil 220.0 30800.0 212.3 27263.7379 4B 229.0 204.6382 4B 253.0 185.6383 4B 137.0 123.0385 M20 134.0 7580.0 140.6 10365.6387/787 Mil 266.0 243.0388 Ml 5 126.5 17710.0 73.5 6985.2388 M20 81.0 11340.0 75.3 8722.2388 4B 139.0 124.2391 M20 58.0 8120.0 58.2 7881.8393 Ml 3 125.0 17500.0 120.7 14549.5398 Ml 5 139.0 19460.0 146.6 18437.8401 M15 81.0 11340.0 78.6 10629.1402 M20 81.0 11340.0 74.9 10266.4405 4B 46.0 36.2411 4B 196.0 169.4787 Mil 183.5 58.5BATES 4B 46.0 22.4CHARGEX 4B 24.0 10.9CRAWLEY 4B 35.0 26.2FPP 4B 11.0 5.0HARSHMA 4A 23.0 9.3HRS 4B 163.0 12.0 45.0 11.9MRSDEV 4B 24.0 18.7OTTEST Mil 47.0 73.2PN 4B 105.0 55.2POLTEST 4A 46.0 64.5 31.7 59.3PPGAMMA 4B 46.0 29.8TISOL 4A 0.6TUNE Mil 11.0 600.0 8.9 250.1TUNE M13 150.0 134.7See Appendix C for experiment title and spokesman.92TM41q modes. The major source of excitation of these modes is a net top-to-bottom dee tip voltage caused by nonuniform sagging of the resonator strongbacks. Since the TM31q mode is symmetric across the centre, while the TM410 mode is antisymmetric, these two modes are excited with different amplitudes depend­ing on tip misalignments along the dee gap, with the TM310 mode being more strongly excited. Calculations showed that ohmic heat­ing due to rf leakage alone cannot account for the observed strongback temperatures. The rf leakage voltage and frequency, and strong- back gap separations have been shown to be appropriate for multipactoring to take place, which in fact is the mechanism responsible for the strongback heating. This is consis­tent with the observation that the strong­backs become hotter as the leakage voltage is lowered into the multipactoring range. To measure the leakage, 17 rf probes have been installed in quadrants 3 and 4. Each probe signal is rectified in a diode detection system and connected to a 14-bit scanning CAMAC ADC channel. The inherent nonlinearity of the diode peak detector has been corrected for in the acquisition and display software, and is about 0.5 dB over a 15 dB range. Accurate calibration of the leakage probes including the cables permits the measurement with a precision of ±3%. Curve-fitting routines have been incorporated into the acquisition programs to extract the relative contributions of the TM310 and the T M ^ q modes to the overall leakage. It was con­firmed both in the model and with the machine measurements that the TM31Q mode is the major contributor to the leakage. By manipulating combinations of the ground arm and hot arm alignments, particularly in segments 7 and 8, it is possible to change the voltage profile of the dee gap to mininize the excitation of either mode. Experiments in the model showed that the leakage can be reduced by as much as 30 dB. In the machine the number of adjust­ments are more restrictive, but an improve­ment of 20 dB was obtained. After the October shutdown the resonator ground tips were sub­stantially tuned such that the rf leakage voltage in most of the beam gap region is below the multipactoring threshold. The leakage profiles before and after the shut­down are shown in Fig. 96, indicating reduc­tions of an order of magnitude in some cases.Because of the sensitivity of the leakage field to ground arm tip adjustments, a system to control the tip tuning motors has been designed and is being implemented to provide more automated tuning of the resonator sys­tem. The system will also be used to tune the cavity for both the fundamental and third harmonic resonances. It will provide tools to reduce the number of variables which have an effect on the tuning requirements. The initial system uses an IBM PC to display variables and assign 8 shaft encoders for manual control of various combinations of resonator tuning motors. The system can be easily reprogrammed in a high-level language as new information becomes available.New rf systems planned for installation at TRIUMF include a third harmonic flat-topping system, a fourth harmonic booster cavity and an rf extraction deflector operating at the11.5 MHz subharmonic of the main frequency. A new modular rf control concept is being adopted to develop the basic building blocks for each of the required rf control systems. The front panel layout with analogue test outputs has been designed. The EUROCARD format has been adopted for construction of the control modules. One of the main design considerations is that all parameters are to be accessible and controllable by an external central computer. The central access will permit the utilization of expert systems and adaptive control techniques for remote debug­ging and loop parameter optimization. In order to carry out diagnostics on individual rf systems it is necessary that each control system also be operational in a local manual mode, independent of the central computer. 'Soft knobs' have been developed to permit bumpless transition between remote and local control. Construction of a prototype rf con­trol system is nearing completion and will be tested on the rf test facility. A new feed­back control module has been designed andresonator numberFig. 96. RF leakage in the beam gap, before the fall shutdown and after reduction by ground arm tip tuning.93regionFig. 97. Effect of movement of standard (#3) segments and new (#4) segments on the reso­nant frequencies of the fundamental and third harmonic waveforms. The straight line corre­sponds to the exact ratio 3:1.incorporated into the fundamental rf voltage control system of the cyclotron, which allows the addition of an arbitrary error signal into the feedback loop. To test the module a signal derived from the beam time of flight was injected into the feedback loop and has reduced the variations in the time of flight caused by voltage variations in the gap voltage by 50%.Studies have been carried out of the effect of changing the pivot point of the ground arm deflections on both the fundamental and third harmonic frequency shifts. Both SUPERFISH and transmission line approximation calcula­tions show that by flexing the ground arm at 1.25 m from the tip, it is possible to change the fundamental frequency without affecting the third harmonic frequency. This design has been incorporated into the 8 newly installed segments. It was found that a value of 3.0 for the frequency ratio f3/fl could be achieved by moving the #4 ground arm panels. The effect of tuning different ground arm segments is shown in Fig. 97. Using a double pulsing system, and with the achieved 3.0 ratio, fundamental and third harmonic power were simultaneously coupled into th