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

Annual report scientific activities, 1984 TRIUMF Jun 30, 1985

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

Item Metadata


51833-TRI-Annual Report-1984.pdf [ 116.19MB ]
JSON: 51833-1.0107784.json
JSON-LD: 51833-1.0107784-ld.json
RDF/XML (Pretty): 51833-1.0107784-rdf.xml
RDF/JSON: 51833-1.0107784-rdf.json
Turtle: 51833-1.0107784-turtle.txt
N-Triples: 51833-1.0107784-rdf-ntriples.txt
Original Record: 51833-1.0107784-source.json
Full Text

Full Text

TRIUMFANNUAL REPORT SCIENTIFIC ACTIVITIES 1984MESON F A C I L I T Y  OF:UNI VE RSI TY  OF ALBERTA SIMON FRASER UN IVER SITY UN IVER SITY OF VICTOR IA  UNI VE RSI TY  OF B R I T I S H  COLUMBIAOPERATED UNDER A CONTRIBUTION FROM THE NATIONAL RESEARCH COUNCIL OF CANADATRIUMFANNUAL REPORT SCIENTIFIC ACTIVITIES 1984TRIUMF4004 WESBROOK MALL VANCOUVER, B.C. CANADA V6T 2A3EXISTINGPROPOSEDPROTON HALL EXTENSIO NREMOlHANDIFACILI'r SERVICE BRIDGESERVICEA N N E XEXTENSIONH POL; IO Nr'ELINGfYU p ) d i  idCHEMISTRYA N N E X 42  MeV  ISOTOPE PRODUCTION  CYCLOTRONM ESO N  HALLM ESO N  HALL EXTENSIONM 9 ( tt / j j ) 2 0 ( jj )INTERIM  RADIOISOTOPE  LABORATORYIZEDOURCEBAT HOBIOMEDICALLABORATORY1NEUTRONACTIVATIONANALYSISo = = = W - ™ - - - - ^THERMALNEUTRONFACILITYM ESO N  HALLSERVICEA N N EXFOREWORDThe depth and breadth of the scientific program in 1984 as recorded in this annual report of scientific activities is extremely gratifying to those of us who can remember the be­ginning of this project nearly twenty years ago. The more than fifty experiments reported in this document utilize the wide spectrum of particle beams available from this unique facility. Although some of the areas under investigation in the experimen­tal program were anticipated many years ago, it is very pleasing to see that a large number of the current experiments were not anticipated and it is obvious that the strength of the research program lies partly in the fast reaction time to important new events in the developing fields of science.1984 has also seen the development of several new important facilities that augur well for future science programs. In the proton hall the new development of the medium resolution spec­trometer, which required some heroic efforts from many staff members, has been commissioned and the results are extremely 8ratifying. Similarly the new M15 channel was commissioned in record time. This channel will support the rapidly developing fields of science in which muons are used to probe liquids, solids and gases, inquiring about a wide range of questions.Due in great part to the exuberance, good will, and the many capabilities of the current Director, the staff at TRIUMF has, in general, shown a very high morale. The camaraderie and high morale are obviously reflected in the high performance level of the various groups within the TRIUMF organization.The cyclotron itself has performed extremely well in 1984 and many new ideas for experiments have emerged. Excellent service has been given to our participating scientists from all parts of the world. I can only hope that the continued cohesiveness at the site will express itself in many more years of fruitful scientific endeavour.Although Dr. John Webster has stepped down from his position as Chairman of the Board, he continues to be a much respected member of the Board and I commend him for his excellent record of service for TRIUMF.G. Croydon NeilsonChairman of the 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 385 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 400.CONTENTSINTRODUCTION lSCIENCE DIVISION 3Introduction 3Particle Physics 3Photon asymmetry in radiative muon capture in 't0Ca 5Search for muon electron conversion y~+Ti -*■ e~+Ti using the time projection chamber ^Test of charge-symmetry in n-p elastic scattering at 480 MeV 8Measurement of the slope of the tt° electromagnetic form factor 10Measurement of muon decay asymmetry parameter 6 10A study of the tt+ * e+ve decay 12Radiative muon capture on hydrogen 13Test of charge symmetry by a comparison of 7r~d > nn with ir^ d pp 14Branching ratio of ir° e+e" 17Nuclear Physics and Chemistry 19Complex reaction mechanisms 19p+160 elastic scatteringBroad pionic X-rays 19Radiochemical study of ot(E) for 209Bi(p O 210-xAt from threshold to 800 MeV 21Coincidence studies of it1 absorption on 3He at 85 MeV 22A study of (p,n) and related reactions 25Proton-proton bremsstrahlung 26In search of a tredecabaryon resonance 27Negative pion absorption at rest in light nuclei 27Pion production from 12C and 10B with polarized protons of 350 MeV 30Search for evidence of delta-nucleus interaction intermediate state in proton elastic scattering 30The (p,2p) reaction and the momentum distribution of the deuteron 31Inclusive scattering of pions from very light nuclei at 100 MeV 32Studies of the A(p,tt“)A+1 reaction 33Energy and angle dependence of the 6Li(ir+, 3He)3He reaction 33Pion double charge exchange on 180 35Transverse spin-flip probabilities in 2‘tMg(pp’) and 't8Ca(p,p') 36Research in Chemistry and Solid-State Physics 37Formation and reactivity of positive muons and muonium in gases 37Muonium in ice 40Muonium chemistry in condensed media 41Amorphous spin glasses 41Muonium relaxation on silica surfaces 42The chemistry of pionic hydrogen atoms 42A comparison of muonic molecule formation rates in HD to H2+D2 gases 44Giant muon Knight shifts 46Muon spin relaxation in random spin systems 47Temperature dependence of muonium reaction rates 49The magnetic superconductor YgCo7 50Muon spin rotation studies of platinum catalysts 51The reaction of muonium with hydrogen peroxide in water 51Muonium-radical formation mechanism 52Search for superdiffusion in stressed vanadium 52Quantum diffusion of muons and muonium 52Positive muon probing solitons in polyacetylene 53Muonium addition reaction in the gas phase 53vii5656Theoretical Program IntroductionNucleon-nucleon potential and applications to finite nuclei ^Bremsstrahlung and off-shell processes Relativistic effects in nuclear physics Nuclear reactions and electromagnetic effects Mesonic and antiprotonic atoms Pion-nucleus physics The quark model and QCDHigh energy physics and weak interactions Field theory Muon spin rotationAPPLIED PROGRAMS DIVISIONRadioisotope processing (AECL)500 MeV radioisotope production facilities Positron emission tomography (PET)TRIM programCYCLOTRON DIVISIONOperational servicesVacuum Remote handlingEXPERIMENTAL FACILITIES DIVISION585960 6263646768 687070Introduction ^Biomedical program ^42 MeV cyclotron ^7677 808282Introduction ggBeam production ggCyclotron ggCyclotron developmentRF system ggAlternative extraction system ^Cyclotron engineering physics ^Probes and diagnostics Ion sources and injection system Primary beam linesThermal neutron facility (TNF)Control system100969899 99Plant services Power supplies101104104Introduction gggExperimental support ^ggData acquisition systems and CFATNucleonics and IAC ^gyDetectors facility 10gMWPC facility 108Meson hall ^ggMil channel 109Ml3 low energy ir-y channel 1QgMl5 channel ^ 2QQD spectrometer ^ 3M20 channelBeam line IB 215Beam line 2CviiiEXPERIMENTAL FACILITIES DIVISION (cont'd)Proton hall H 3(p,n)(n,p) facilityNeutron facility 116Beam line 4BMRS ll7Targets H ®Experimental facilities engineering 120ACCELERATOR RESEARCH DIVISION 122Introduction 122Beam development 123Cyclotron 123Primary beam line 12®-i O 1Secondary channels IOOBeam line diagnosticsComputing services 133KAON factory studies 136TECHNOLOGY AND ADMINISTRATION DIVISION 146Introduction 146Site services 146Safety program 146Building program 14°Mechanical engineering I50Design office 1^1PlanningControls, electronics and computing 132Introduction 1321 50ElectronicsMicrostructuresData analysis centre 133issAdministrationBusiness Office 133Visitor housing 137CONFERENCES, WORKSHOPS AND MEETINGS I58ORGANIZATION I60APPENDICESA. Publications 163B. Users groupC. Experiment proposalsixINTRODUCTIONThe year 1984 was for TRIUMF a year of record beam delivery and science output and, equally significant, of facility developments which herald the next decade of science at TRIUMF.There may not be any other major facility in the network of large North American facili­ties for subatomic physics at which such a large fraction of the year is devoted to beam delivery. In more than 40 weeks of operation - more than two-thirds with high intensity and the remainder with polarized beam - the beam was delivered for 88.5% of the scheduled time. The total number of protons delivered during the year (322,000 pAh) was about a third more than the previous best record, established last year.Although in the whole life of an accelerator project the best measure of its accomplish­ment is the sum of the science carried out during periods of beam delivery, in a young project such as TRIUMF much of the develop­ment and a significant fraction of the year's highlights occur during the accelerator shut­down periods. This was certainly true for1984. For example, the magnificent commissioning of TRIUMF's new surface muon channel (M15) occurred during the September shutdown of only a few weeks. The intense and highly organized activity during this short time period included removal (and sub­sequent replacement) of the massive shielding blocks for access to the production target (1AT1), the installation and accurate align­ment of the many large magnets and other beam line elements, the installation of vacuum pipes and couplings, and the connection of water and electrical services. The burst of effort culminated in beam delivery in M15 after a commissioning period of weeks rather than the anticipated six months. The accom­plishment is a tribute to the efforts of the group led by Dr. David Garner and the high performance of the channel to the design ef­forts of Dr. Jaap Doornbos and his colleagues.The hallmark of TRIUMF for 1984 will be the commissioning and initial operation of the medium resolution spectrometer (MRS). This spectrometer had been a long time in the design and building stage, but its time for centre stage had finally arrived. Driven by new worldwide interest to have a spectrometer of good resolution (100 keV) in TRIUMF's unique and important energy interval (180- 520 MeV), the final commissioning of MRS wasgiven the highest project priority in the first half of 1984. The groundwork for a dispersion-matched spectrometer-plus-beamline system had been laid in 1983 with the instal­lation of the twister in beam line 4B. What was needed last year was a proper focal plane detector and much attention to the tuning and alignment of the entire system from the point of cyclotron extraction. The tireless and highly competent efforts of a large group led by Dr. C. Andrew Miller and Dr. Otto Hausser achieved 100 keV in mid-year. This is a major milestone for a system which already benefits from high momentum acceptance (Ap/p = 10%) and TRIUMF's excellent beams and which short­ly expects to see the addition of charge ex­change [(p,n) and (n,p)] capabilities and the installation of focal plane polarimeters. For at least a decade this should prove to be one of the world's most important facilities for nuclear physics, as manifested by the backlog of EEC-approved experiments waiting for its use.One of TRIUMF's largest and most important experiments (charge symmetry breaking, Exper­iment 121) finished its data production runs at the end of the year. This experiment is a second-generation one in the nucleon-nucleon program of TRIUMF. Over TRIUMF's first seven years (1976-1982) the major BASQUE program mapped out the properties of the nucleon- nucleon force over intermediate energies. This constituted a step-function change in our knowledge of the strong interaction between nucleons. Using this first-genera­tion program as a platform the second genera­tion of experiments looks for very small but important asymmetry effects in the force. In Expt. 121 one looks for minute differences in the force (one part in a thousand) if one interchanges all neutrons and protons. The property of the force under which it is very nearly invariant under such a replacement is called charge symmetry. Theory predicts breaking of this symmetry by a few parts per thousand. The experiment looks at the scat­tering of polarized protons from neutrons (and vice versa) for angles at which the analysing power crosses zero. To achieve the desired goal the zero crossing must be deter­mined, in each case, to an absolute accuracy of a twentieth of a degree. The experiment required years of planning and about a full year of running by a large team under the able leadership of Prof. W.T.H. van Oers. It also required several technical miracles.First, the construction and operation of a large volume frozen-spin target (North Ameri­ca's first) at one fifteenth of a degree above absolute zero. TRIUMF has a highly cap­able cryogenic target group led by Dr. Dennis Healey, which built this and many other low temperature targets. Next, the polarized ion source group led by Dr. Paul Schmor more than doubled the anticipated polarized beam inten­sity delivered on target to the experiment. For many days intensities approaching 1 pA on target were achieved.In the meson hall a large number and variety of experiments benefitted from the increased beam delivery. The muon conversion experiment is approaching its goals. So far none of the sought-for neutrinoless muon events have been observed, but the continued fine operation of TRIUMF's time projection chamber for this ex­periment has lowered the limit to a value (approaching 10-11) which places important constraints on theory. The pSR program is mushrooming - using the M20 channel commis­sioned in 1983 and the newly commissioned M15 channel - and has demonstrated an important new level-crossing technique. A great deal of interest centred on the pion double charge experiments at 50 MeV and also on the pion- deuteron experiment for tensor polarization parameters.In TRIUMF's applied program the year-round operation of the CP-42 cyclotron at high in­tensity benefitted both the commercial pro­duction of isotopes and the production of isotopes for our PET (positron emission tomography) facility. In pion therapy thenumber of patients treated (69) increased significantly and produced some very encour­aging results.TRIUMF major future plans are for a KAON fa­cility in which the present TRIUMF cyclotron would be used as an injector into a new sys­tem of accelerators which would boost the energy sixty-fold while retaining the high current. Such a facility would provide im­portant new beams of kaons, neutrinos, anti­nucleons and many other particles. In 1984 the accelerator concepts for this purpose received a great deal of study and the pro­posal for the facility neared final prepara­tion. The KAON facility would be Canada's piece in the world network of large new ac­celerator facilities for subatomic physics. Interest in the whole network is inspired by the recent dramatic developments in our understanding of quarks, leptons and their interactions. With the KAON facility Canada will attain to a world-leading position in the field for the 1990s.In the management of TRIUMF the chairmanship of TRIUMF's Board of Management passed from the very capable hands of Dr. John Webster to Dr. G. Croydon Neilson who has been one of TRIUMF's strong leaders since the beginning of the project. The whole development of TRIUMF continues under the strong encourage­ment provided by the National Research Coun­cil which funds the operation of the facility. Especially important are the direct interest of Dr. Paul Redhead, Chairman of the NRC Advisory Board on TRIUMF (ABOT), and of the NRC liaison officer, Dr. Jeff Child.2SCIENCE DIVISIONINTRODUCTIONThe year was an excellent one for the Science Division, mainly because the cyclotron ran so well. Over 300,000 pAh of unpolarized beam and average currents of more than 300 nA of polarized beam set new records for the proj­ect. In all 51 experiments received beam, with 20 of them completing data-taking in1984. The facilities available to experimen­ters were substantially improved. The MRS magnetic spectrometer upgrade project finally achieved energy resolution better than 100 keV, the culmination of a magnificent ef­fort by scientists, engineers and technicians from both Science and Experimental Facilities Division. The spectra produced by this in­strument are cleaner and have less background than those coming from any other comparable spectrometer. Further improvements and addi­tions are planned, including the installation of a focal plane polarimeter to measure the polarization of the detected protons. This will be installed and commissioned early in1985. Another major development has been the construction of the components for a facility for studying (p,n) and (n,p) reactions in nuclei using the MRS as a detector. This facility will also be commissioned in the first half of 1985. It is clear that the MRS will be the heart of our effort in proton physics over the next few years, and everyone connected with the successful upgrade of this device is to be congratulated.The other major addition to experimental facilities was the installation and commis­sioning, in record time, of the M15 surface muon line. This beam line was being used for pSR experiments within a few weeks of the September shutdown when the major components were installed.In experimental physics there were a number of highlights. The charge symmetry breaking experiment finished all its data-taking, with just a short background run spilling over into January 1985. The completion of this measurement has necessitated major efforts from both the experimenters and the labora­tory itself, particularly in building the frozen spin target and in providing a pola­rized proton beam of enough intensity. The p+e conversion experiment being carried out in the time projection chamber had a very successful year, more than doubling the num­ber of muon stops observed in previous years and setting a new upper limit on the branch­ing ratio. This experiment, the largest at TRIUMF at the present time, is expected to finish data-taking in the summer of 1985.The Berkeley-Northwestern-TRIUMF group which carried out the precision measurement of the muon decay polarization parameter £ reported in last year's Annual Report used the same apparatus to make an improved measurement of the anisotropic shape parameter (5) in muon decay. Data-taking for the experiment was completed in January and preliminary results reported in the summer. The anticipated pre­cision is a factor of three improvement over previous measurements and will set limits on the mixtures of scalar, pseudoscalar and tensor couplings in the weak interaction.There were two major runs with the QQD pion spectrometer in 1984, the most important re­sult of which was a measurement of the angu­lar distribution for double charge exchange on 180 at 50 MeV. This experiment complements the double charge exchange measurements made on 11+C with the TPC last year. The combina­tion of the Ml3 low-energy pion line and the QQD spectrometer is well suited to these ex­periments, which may be expected to continue to be an important component of the QQD program.In addition to the success with M15 the high­light of the year in the pSR community at TRIUMF was the demonstration that the dc sep­arator in M20 can be used as a spin rotator to rotate the muon polarization so that it is transverse to the momentum. A longitudinal magnetic field, into which surface muons may be injected much more easily than into a transverse field, can then be used to precess the muon spin. The M20 spin rotator facility is presently unique in the world.As usual, meetings of the Experiments Evalua­tion Committee were held in July and Decem­ber. In July 23 experimental proposals were received and in December, 39. These numbers are to be compared with an average of 12-15 proposals received only two years ago. There appears to be a shift in emphasis away from M13 and toward Mil, the high-energy pion channel. The phenomenal growth in proposals3submitted to TRIUMF can be taken as a sign of the vitality of the project and the competi­tiveness of the beams and facilities avail­able at the laboratory.Finally, mention should be made of three workshops receiving TRIUMF support:Radiative Processes in Few-Nucleon Systems The TRIUMF-ISOL Workshop High-Energy Spin PhysicsThe TRIUMF-ISOL Workshop was held at Mont- Gabriel, Quebec and organized by Simon Fraser and McGill Universities to examine the possi­bilities of using TRIUMF as the site for a high performance on-line isotope separator. There is no doubt that this could be done to produce a facility which would be unique in many respects, and a detailed proposal isbeing prepared. The Workshop on High-Energy Spin Physics was held in conjunction with the Western Regional Nuclear Physics Conference at Lake Louise, Alberta. It is hoped that this will be the first of a series of winter workshops at Lake Louise.Running schedule for 1985Jan 3-Jan 10 10 pA unpolarizedJan 11-Feb 5 polarizedFeb 6-Mar 19 shutdownMar 20-May 22 high intensity unpolarizedMay 23-Jun 15 polarizedJun 16-Aug 21 high intensity unpolarizedAug 22-Sep 14 polarizedSep 15-0ct 22 shutdownOct 23-Dec 2 high intensity unpolarizedDec 3-Dec 23 polarizedThe 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 147Photon asymmetry in radiative muon capture in 40Ca (M.D. Hasinoff, UBC)The value of the pseudoscalar coupling con­stant, gp, for a nucleon bound inside a nuc­leus is predicted to be sensitive to explicit mesonic degrees of freedom in the nucleus. Since most ordinary capture experiments in nuclei are quite insensitive to gp the exper­imental work has concentrated mostly on mea­surements of inclusive radiative muon capture where one can observe both the gamma energy spectrum and the angular correlation between the photon and the muon spin direction. The gamma asymmetry is predicted to be the ob­servable which is least sensitive to the uncertainties in the nuclear models.The experimental details of our TRIUMF exper­iment were reported in last year's Annual Report and these will not be repeated here. As mentioned in our earlier report one source of prompt ir-induced background due to the forward leg of the M20 channel could not be totally eliminated. Thus our gamma data between 55 and 90 MeV are contaminated with a 43 ns background component. For our contri­bution to the Heidelberg Conference we aver­aged the early time background and then sub­tracted it from the proper time region before fitting the gamma asymmetry. This produced an asymmetry value with an uncertain error, since any possible correlation between the asymmetry parameter and the background termwas ignored. Thus we have since spent a good deal of time developing the proper software to include an empirical background term in the MINUIT fitting program. Figure 1 shows the gamma asymmetry spectrum after subtract­ing both the fitted lifetime and empirical background terms. The value of the gamma asymmetry over the energy interval 55 MeV < Ey < 90 MeV is 0.127±0.037.The integral gamma asymmetry is shown in Fig. 2 as a function of the lower cut-off en­ergy. Below 55 MeV the gamma data are contam­inated by bremsstrahlung from the decay elec­trons and so the integral asymmetry quickly becomes negative below 50 MeV.We have also developed a detailed Monte-Carlo program (based on the EGS program) to deter­mine our electron and gamma response func­tions and to provide a value for the residual y- polarization at the moment of capture. This Monte-Carlo program includes the elec­tron bremsstrahlung and Compton scattering processes in our calcium target as well as in all materials between the target and TINA.Figure 3 shows our measured e+ and e“ differ­ential asymmetries along with the normalized Monte-Carlo predictions. The depolarization of the y~ during the cascade is determined to be 0.15±0.01, which is consistent with the0.2  0 .3  0 .4  0.5 0 .6  0 .7  0.8TIME ( / is )Fig. 1. Gamma asymmetry spectrum for the en­ergy interval 55-90 MeV. The contributions from the fitted lifetime and the empirical background have been subtracted from the original data.>crUJ>-<U Jcs><crU J£ 0  M eVIN T E G R A L /A S Y M M E TR Y */ A (E y )dEy ^ c u t o f fi i "z 130  4 0  50  6 0  7 0  80ENERGY CUT OFF (MeV)Fig. 2. Integral gamma asymmetry as a func­tion of the lower cut-off energy.5E e-  (M eV)Fig. 3. Measured e+ and e“ differential en­ergy curves. The Monte-Carlo curves have been scaled by 0.60 and 0.09 as indicated.factor of 1/6 quoted in the literature. How­ever, the p+ polarization is less than expec­ted. Monte-Carlo calculations of the effects of multiple scattering and the finite angular acceptance of our apparatus are currently under way.Experiment 104Search for muon electron conversion +  Ti-^e +  Ti using the time projection chamber (D.A. Bryman, Victoria/TRIUMF)A search is being performed for the neutrino- less lepton flavour violating reactionp~+Ti -*■ e“+Ti (1)In the coherent process the nucleus remains in the ground state resulting in emission of a single electron with energy Ee = my-B = 104 MeV where my is the muon mass and B is the muon binding energy. Incoherent muon- electron conversion resulting in an excited nucleus is suppressed by Pauli blocking.The experiment is being performed using the TRIUMF cloud muon beam line M9. The 73 MeV/c beam passes through an rf particle separator and is degraded and stopped in a 2 g/cm2 shredded titanium target (density 0.1 g/cm^) at the rate of 106 p~/s. Surrounding thetarget is the electron detection system con­sisting of a time projection chamber (TPC) and two sets of trigger counters as shown in Fig. 4. The TRIUMF TPC is a large volume at­mospheric pressure drift chamber operating with parallel electric and magnetic fields. Some parameters of the TPC system are listed in Table I. The central high voltage plane causes ionization electrons from charged par­ticle tracks to drift to either end-cap where they are detected by an array of proportional wires. Independent x, y and z co-ordinateFig. 4. A perspective view of the TPC. The TPC has 12 sectors with 12 anode wires per sector separated radially by 2.54 cm. The innermost wire is at a radius of 19 cm. The maximum drift length to the central high voltage plane is 34 cm. The numbered elements are: (1) the magnet iron, (2) the coil, (3a) and (3b) trigger scintillators, (4) outer trigger proportional counters, (5) end-cap support frame, (6) central electric field cage wires, (7) central high voltage plane, (8) outer electric field cage wires, (9) in­ner trigger scintillators, (10) inner trigger cylindrical proportional wire chamber, and (11) end-cap proportional wire modules for track detection.6Table I. TPC parameters. Table II. Background estimates.GasMagnetic field Drift field Drift velocity Anode wire gainAr(80%), CHy(20%) at 1 atm 9 kG 250 V/cm 7 cm/ys 5 x 104ProcessEquivalent branch­ing ratio levelinformation is obtained for each of up to twelve track segments. The anode wire posi­tions and the induced charge distributions on the segmented cathode pads give the y and x positions, respectively, and the drift times relative to the trigger determine the z co­ordinates. The inner trigger counter set consists of six scintillators and a cylindri­cal MWPC and the outer set is made up of six planar MWPCs each sandwiched between scintil­lators, as shown in Fig. 4.The trigger is enabled for a period 20 to 600 ns following a stop signal, the coinci­dence of four beam scintillators preceding the target in anticoincidence with veto coun­ters which surround it. The first level trigger requires a coincidence of both inner counters and at least two outer counters. In order to eliminate space charge effects due to leakage of positive ions from the propor­tional wire region into the drift region of the TPC, a dual grid structure blocks all ionization from reaching the end-cap wires except when a first level trigger signal is present. For the second level trigger at least six TPC wires are required to fire in the appropriate sectors of the TPC.The acceptance and performance of the detec­tion system is monitored by studying the decay ir+ ■* e+ve. Pions are stopped in the target and the 70 MeV/c positrons are ob­served in the TPC with the magnetic field lowered to B = 6 kG so that their curvatures match those of electrons at 104 MeV/c when B = 9 kG. The acceptance of the detector system is approximately 20%. The momentum resolution for electrons at 104 MeV/c is ex­pected to be approximately 4 MeV/c (FWHM), based on Monte-Carlo calculations which use position resolution data derived from cosmic- ray tests.Potential sources of background for coherent muon-electron conversion in the present sys­tem are listed in Table II along with their estimated levels. Pion contamination in the beam could produce single 100 MeV/c electrons via radiative capture and subsequent asymmet­Beam pions Cosmic raysy- decay in atomic orbit Radiative muon capture-10-10-13- 1 2- 10“ 12 <10“ 12ric conversion. Pions are suppressed by an rf separator and by a subsequent range tele­scope from an initial ratio it/y = 1 to ir/y < 10 at the experimental target. Adequate re­jection of the remaining pion-induced events is obtained by eliminating any event candi­date which is in prompt coincidence with a beam particle. Cosmic ray-induced backgrounds have been studied during beam-off periods and have been substantially suppressed by large drift chambers which surround the TPC magnet. The expected levels of background from radi­ative muon capture and from bound y- decay are derived from Monte-Carlo calculations.During an initial running period Ny = 3xl012 muons were captured in the Ti target. After all cuts were applied the electron momentum spectrum shown in Fig. 5(a) was obtained. This spectrum is consistent with bound y- decay. Figure 5(b) shows a calculated spec­trum for the coherent reaction (1) with a branching ratio of 10-10. Using the data ofELECTRON MOMENTUM (MeV/c)Fig. 5. (a) The observed electron spectrum,(b) The dashed lines represent a Monte Carlo generated spectrum for y“+Ti e“+Ti with a branching ratio of R - 10-10.7Fig. 5(a) a preliminary upper limit (90%C.L.) for the branching ratio is:r(p"+Tl e“+Ti)R = — —r(y“+Ti + vu+...)NUF < 2 x 10-11where T  = 3.5% is the effective acceptance for this data set. The experiment is contin­uing.Experiment 121Test o f charge-symmetry in n-p elastic scattering at 480  MeV(W.T.H. van Oers, Manitoba)This calendar year saw four long data-taking runs February 24-March 9; August 4-22; Sep­tember 27-October 17; and December 5-24. In past annual reports we highlighted instrumen­tation design, construction and testing; a description was also given of a series of fourteen test runs. With the data-taking phase of this test of charge symmetry almost completed, we present here a short descrip­tion of the experiment.The experiment measures the difference AA between the neutron and proton analysing powers An and Ap in n-p elastic scattering at 480 MeV. Designed as a null-measurement, requiring no accurately known polarization standards, the experiment determines the dif­ference in angle at which An and Ap cross through zero (at 71° c.m. or 32° lab). The two interleaved phases of the experiment consist of scattering polarized (unpolarized) neutrons from an unpolarized (polarized) tar­get of the frozen spin type. The experiment is to provide an unambiguous test of class IV charge-symmetry breaking effects to the level of AA ~ 0.001, corresponding to a laboratory angle difference at the zero-crossing of -0.04°.A 480 MeV polarized neutron beam is produced via transverse polarization transfer in the D(p,n)2p reaction at 9° lab using a 20 cm long liquid deuterium (LD2) target. A super­conducting solenoid rotates the transverse polarization direction of the 500 MeV proton beam by 90° clockwise into the horizontal reaction plane. Polarization direction re­versals are implemented at the polarized ion source. The neutron beam passes through a3.37 m long lead collimator. In the neutron scattering area the horizontal component of the neutron polarization is precessed into the vertical plane by two dipole magnets (see Fig. 6). The component of polarization per­pendicular to the reaction plane is precessed simultaneously along the direction of travel of the neutron beam. A proton beam polariza­tion with typical values in the range 0.65 to 0.70 produces a neutron beam polarization of ~0.50. In the interleaved second phase of the experiment a neutron beam with zero transverse polarization components is pro­duced as above, but with an unpolarized pro­ton beam incident on the LD2 target.The proton beam polarization is monitored continuously by a polarimeter observing p-p scattering at 17° lab using a left-right symmetric detector telescope arrangementProfile MonitorFig. 6. Experimental layout of a test of charge symmetry in n-p elastic scattering at 480 MeV.complete with recoil counters. The proton beam energy is monitored using six element range counters following the forward angle (17° lab) polarimeter detector telescopes. Two sets of split-plate secondary electron emission monitors (SEMs), coupled via a feed­back system to two sets of steering elements located upstream in the beam, line allow the centroids of the proton beam to be locked into position to within ±0.15 mm. After traversing the LD2 target the proton beam is deflected away from the neutron beam and transported to a beam dump.The collimated neutron beam is incident on a frozen spin polarized hydrogen target (FST) with a target cell 40 mm in height and a diameter of 40 mm containing 1.6 mm diameter butanol beads. In the first major data-tak­ing run operational temperatures were 75 mK. The average positive polarization was +0.66 while the average negative polarization was -0.72. In the second data-taking run opera­tional temperatures were about 60 mK with average positive and negative polarizations of +0.81 and -0.87, respectively. In the third data-taking run the FST performed simi­larly. The strength of the holding field is 0.257 T at the FST causing an average deflec­tion of 1.2° for the recoil protons. A com­plete measurement cycle lasts four days and consists of holding field directions up and down each with positive and negative target polarizations for incident unpolarized neu­trons, interleaved with the target depol­arized for incident polarized neutrons. The FST holding field strength is both stable and reproducible to better than ±0.25 mT.After traversing the FST the neutron beam passes a profile monitor. The neutron beam profile exhibits an intensity distribution which is flat-topped over a rectangular area and is reproducible in position to within ±0.5 mm from run to run. The area of uniform intensity at the position of the FST is 56 mm wide by 40 mm high encompassing the target cell.The profile monitor is immediately followed by a polarimeter determining left-right and up-down asymmetries. The effective analysing powers of the neutron polarimeter were de­rived from the amplitudes of the precession curves. The zero crossings of the precession curves provided the calibration of the spin precession magnets.The neutron energy spectrum consists of a relatively narrow peak (calculated FWHM =11 MeV) around 480 MeV and a smooth flat background [Amsler et al., Nucl. Instrum. Meth. 144, 401 (1977)]. Approximately one- half of the neutrons are contained in the peak corresponding to an intensity of 9 x 10° s-1 cm-2 per 100 nA of proton beam. The polarized proton beam has intensities greater than 300 nA for each of the three spin states and reached intensities of 650 nA during one week of running in late 1983. A detailed description of the neutron beam facility is presented elsewhere [Abegg et al., Nucl. Instrum. Meth., in press].Scattered neutrons and recoil protons origin­ating in the FST are detected in coincidence in two left-right symmetric detection systems allowing cancellation of many of the system­atic errors to first order. Each neutron detection system placed at an angle of 32° consists of four elements: A neutron scin­tillator array, a plane of charged particle veto scintillators covering the forward side of the neutron scintillator array, a set of small scintillators ('button' counters) used for monitoring on line the gain of the photo­multipliers of the neutron scintillator array by triggering on fast protons, and a charged particle time-of-flight (TOF) scintillator.Each proton detection system is supported by a boom placed at the complementary angle of 51°. The boom carries three functionally distinct systems: a TOF system for determin­ing proton velocity consisting of a TOF start scintillator close to the target and an ener­gy scintillator E which serves as the stop, a set of four 0.58 m by 0.58 m delay line cham­bers for track reconstruction, and a range counter telescope for removal of high-energy and low-energy background. A detailed des­cription of the detection equipment is pre­sented elsewhere [Abegg et al., Nucl. Instrum. Meth. (in press)].With the detection system one measures the neutron TOF and position and the proton TOF, trajectory and energy. Measurements of the opening angle and coplanarity of the coin­cident neutron-proton pair discriminate against quasi-free n-p scattering from heavi­er nuclei, e.g. 3He, 4He, 12C and ie0, pre­sent in the FST. Further constraints on the data are cuts on the reconstructed target image, the summed neutron-proton energy, and the neutron-proton transverse momentum. The combined data from the first two runs gave a statistical accuracy on the difference in the crossover angles for the neutron and proton analysing powers of ±0.11°. With the four9data-taking runs we estimate the final sta­tistical error to be ~±0.06°, slightly larger than originally intended. A preliminary re­sult, not completely corrected for systematic error, is not out of line with existing theo­retical predictions.Experiment 21 7Measurement o f the slope o f the n° electromagneticform factor(J.-M. Poutissou, TRIUMF/UBC; A.W. Stetz, Oregon State)Experiment 217 has measured the slope of the ir° electromagnetic form factor ' a ' by care­fully studying the invariant mass spectrum of the lepton pairs emitted in the ir° -*■ e+e“y decay. The invariant mass x is determined fromx = ---[2m^ + 2E.E_ - 2p.p_ cos0lmir°by measuring E+ and E_, the energies of the positron and electron, in two large Nal de­tectors and the opening angle of the pair in two telescopes of three wire chambers each. To be sensitive to 'a' an experiment must detect pairs e+e~ with large x, which in turn implies large opening angles.The main data-taking runs for Expt. 217 oc­curred in May and June. The experimental arrangement sketched in Fig. 7 was used and data were taken at 3 different opening angles between TINA and MINA (150°, 135°, 60°). The relative normalization for these three runs was obtained from the third Nal detector (Sophie), which remained at the same location for the duration of the experiment to monitor the high-energy photons from ir° + 2y and ir-p -*■ yn.A 7 cm diameter liquid H2 flask was used with symmetric windows and vacuum vessel. Low- energy it- at a rate of 400,000/s were brought to rest in the centre of the flask (FWHM = 5 cm). The last two beam-defining counters were made of hydrogen-free material to avoid it0 production at their location.The position resolution of the wire chambers was 1 mm FWHM, and their efficiencies were monitored continuously (typically 98.5%). The stability of the Nal detectors was monitored continuously, and the absolute calibration was derived from the M13 channel magnets us­ing the surface muon edge. The absolute cal­ibration is known to 150 keV.Fig. 7. Experimental set-up in the 150° geometry.Figure 8 shows the two-dimensional plots of Ex of the pair (the total energy) as a function of the invariant mass of the pair in 150° angular conditions. The lower triangular domain contains the good ir° e+e“y events while the upper line contains the ir“p e+e~n events.In the 60° geometry the average x for all good events is very small (<0.3) and the ex­periment should not be sensitive to 'a' while at 150° x varies from 0.3 to 1. One must eliminate the x > 0.8 events due to the over­lap with ir“p * e+e“n events.We collected ~50,000 events at 150°, 60° and ~30,000 events at 130°. We are in the process of reducing the data and recalibrating each run according to the various monitors of sta­bility and efficiencies. P. Gumplinger is in charge of the bulk of the analysis for his Ph.D. thesis.Experiment 24 7Measurement o f muon decay asymmetry parameter S (J. Carr, Berkeley)The Berkeley-Northwestern-TRIUMF group, which has recently completed data-taking on Expt. 185, used the same apparatus to make an improved measurement of the anisotropic shape parameter 6. The method employed was the muon rotation (pSR) technique. Data were taken over a wide range of the reduced posi­tron momentum x with aluminum targets of two different thicknesses and with two different ySR precession frequencies.10i m i i i i r r  | i ii t t  i r n  | i n  r n T T i  pi i i n  i r i i | r t T T2 .  iT r r n  1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1  i 1 i t  t t i i i i t i  1 1 1 1 1 1 1 1 1 1 1 10 . 0 2 0 . 2 3 0 . 4 4 0 . 6 5 0 . 8 6 1, 07XFig. 8. Two-dimensional plot of (total pair energy in MeV) versus x (invariant mass2 of pair) for the 150° geometry.The results are very sensitive to the de­tailed x calibration and to radiative effects.Much beam time was spent in thoroughly study­ing these systematic effects. A preliminary analysis of the asymmetry as a function of x is shown in Fig. 9; both the V-A theory and the left-right symmetric theory predict 6 =3/4, but mixtures of scalar, pseudoscalar and tensor couplings can cause deviations from this value.In the absence of radiative effects the asym­metry is expected to change sign when x = 0.5; however, internal and external bremsstrahlung cause the zero asymmetry point to shift to somewhat lower values of x. Our preliminary result isr r. - i /o + 0.004 (statistical)6 = °-748 * 0.003 (systematic)Ultimately we expect to reduce the combined statistical and systematic errors to the ±0.003 level. Already the error on the pres­ent preliminary result is half that of the best previous measurement [Fryberger et al.,Phys. Rev. 166, 1379 (1968)]. _. Q „ . . . .J ’ /J Fig. 9. Muon spin rotation asymmetry mea­sured in a preliminary determination of the muon decay parameter <5. Where not visible the error bars are smaller than the dots.X  = p(e)/p(e)max11Experiment 248  A study of the n +-*  e +ve decay (T. Numao, TRIUMF/Victoria)The equality of coupling strengths previously known as electron-muon universality is now naturally widened to lepton universality to encompass the t and any subsequently discov­ered leptons. This principle is built into the successful Weinberg-Salam-Glashow (WSG) electroweak model and it leads directly to the requirement for the V-A doublet structure of the lepton sector as well as to the re­quirement for massless neutrinos. Universal­ity is tested most stringently by the mea­surement of the ir+ev/ir-vyv branching ratio. In the lowest order it can be calculated asF(tt ev)R°  r ( tt ->• pv)f^mg(mg-m|)2 f*2 ffmJ(m2-mJ)2 f f• (1.278 x 10-4).The principle of electron-muon universality in pion decay holds under the assumption that the basic interaction current is of the V-A type if f® = f^ J. Early calculation done in the framework of the non-renormalizable uni­versal V-A theory neglecting the effects of strong interactions gave the radiatively corrected resultRr(ir-»-ev + ir+evy) r(ir->-yv + tt+u v y)= R0(l+S)(l+e) = 1.233 *x 10where 5 = -(3a/ir)£n(m,j/me) = -16(a/ir) ande = -0.92(a/ir). Remarkably, detailed gauge theory calculations found R to be nearly identical to the above result. Further cor­rections to R due to pion structure are esti­mated to be at the level of <0.3%.Deviation from universality could be the result of a number of interesting possibili­ties. If a pseudoscalar current were domin­ant in T+eVg decay rather than axial vector current, then the value of the branching ratio would be radically different: R0 = 5.5. Pseudoscalar interactions could be induced by unusual Higgs particle couplings. Unlike many lepton flavour violating processes in which the reaction rate depends on l/m^, the effectof pseudoscalar Higgs coupling in ir+ev decay would be proportional to 1/m^ due to the interference with the dominant axial vector interaction term.The existence of massive neutrinos could af­fect the branching ratio, too. In addition to the test of universality the same experiment is able to give energy spectra for the search of additional peaks due to the existence of massive neutrinos mixing to the electron neu­trino in the mass region of 60-120 MeV. Al­though the TT+ev decay is helicity suppressed, for massless neutrinos this suppression is not applicable to the heavy neutrino case. Therefore, the search for additional peaks in the Tr+ev spectrum is a sensitive test for such mixing.The most precise experiment to date was done recently at TRIUMF, in which the branching ratio R = (■rr-»-ev+Tr>evY)/(iT->pv+7r-»-pvY) was mea­sured to be R = (1.219±0.014) x 10-4. This was in substantial agreement (lcr) with the standard model calculation. In the experi­ment the single largest uncertainty (0.75%) was in the tail correction for the ir->-ev events under the Michel spectrum from ir-y-e below the cutoff energy at 51 MeV. Signifi­cant improvement in knowledge of the ir*ev tail correction would result if the ir-y-e chain background were suppressed by at least a factor of 10-lt (as low as the ir+ev branch­ing ratio), so that the statistical uncer­tainty from the background would be compa­rable to or less than that of ir+ev events.We have developed a new technique of reducing the ir-y-e chain decay background based on the total energy measurement in the beam count­ers. Because of the lifetime difference between pions (t = 26 ns) and muons (t =2197 ns), the timing gate for earlier time suppresses the ir-y-e background by more than a factor of 100. An additional factor is achieved by a cut on the total energy deposi­ted in the target counter. For ir+ev decay events the total energy deposited in the target (and beam counters) is the sum of the kinetic energy of the pion (25 MeV) and a small part of the positron energy. However, for ir-y-e decay events there is an additional muon kinetic energy (4 MeV). Therefore, two peaks are expected in the spectrum of the total energy in the target counter. If an energy window is set at the lowest energy peak (no muon decay) most ir-y-e decay events are removed. This background suppression method not only works for getting the tail12correction but also reduces the background in the search peaks due to massive neutrinos.A preliminary test was done at the M13 chan­nel. A 77 MeV/c positive pion beam was de­graded and stopped in a scintillation-counter target designed to absorb all muons from Tr+>p+Vy decay. A pion stopped in the tar­get decays either to e+ (E = 70 MeV) by ir+->e+ve decay or to P+ by ir+-vp+vp decay followed by p+->-e+w  decay (ir-y-e chain, E = 0-53 MeV). Positrons were detected with a large volume Nal(TJl) crystal TINA which had a solid angle of 0)/4ir = ~5% defined by a tele­scope of plastic scintillators and wire cham­bers .A typical energy resolution of AE/E ~ 6%(FWHM) was observed for the 70 MeV/c ir+ev peak without using wire chamber information.Figure 10(a)shows an energy spectrum of po­sitions during At of 3-25 ns after a pion stop. The effect of the suppression technique is shown in Fig. 10(b) where cuts on the to­tal energy and the pulse shape in the target counter are applied. A suppression factor due to the cuts of ~200 was achieved in the test.Experiment 249Radiative muon capture on hydrogen(G. Azuelos, Montreal)Radiative muon capture (RMC) is well known to be highly sensitive to gp, the magnitude of the induced pseudoscalar coupling constant of the weak hadronic current, particularly at high photon energies. A measurement of the rate for p- radiative capture can also be interpreted in terms of an induced tensor second class current. This process has been studied recently in nuclear targets, namely 160 and it0Ca, both at SIN and at TRIUMF. However, RMC in hydrogen has never been mea­sured before. Unlike RMC in nuclei it does not suffer from any uncertainties related to nuclear structure effects. The main interest in the measurement of RMC in hydrogen is to determine the extent to which one-pion- exchange dominates the pseudoscalar term in the nucleon axial current. The interestingquestion in the nuclear case is whether ornot the pseudoscalar coupling constant is renormalized by the presence of the othernucleons in the same fashion as the axialcoupling constant gA*CHANNEL NUMBER Fig. 10. A positron spectrum (3 ns < t < 25 ns) from the decay tt+ev (a) without and (b) with cuts on the total energy and the pulse shape in the target counter.Figure 11 summarizes the expected dependence of the rate on g and gpt, where gp is the coefficient of the pseudoscalar term which includes a pion pole term, and g is the co­efficient of an additional pseudoscalar term where the pion pole term was removed. The cal­culations, which are fully relatlvistic and which preserve gauge invariance, are based on Fearing's code [Phys. Rev. C 1^^ , 1951 (1980).At present the best value for gp/g^ lot the nucleon comes from ordinary muon capture (OMC) in liquid hydrogen, where it has been determined to 22%. The proposed measurement of RMC in hydrogen using the TRIUMF TPC will improve this to 12-15% initially after 1 to 2 months running time. Expected rates are up to 8 events per day.The TPC is well suited to RMC measurements because of its large solid angle. A y- converter can be placed in the central cylin­drical region to produce e+e_ pairs which can be clearly identified in the TPC, and there­fore distinguished from neutrons and other backgrounds. Measurements of RMC on light nuclei will be carried out first to evaluate backgrounds and systematic errors. Expected event rates are of the order of 100/h.The liquid hydrogen target and converter- scintillator sandwich to be used for measure­ment of RMC on nuclei have been designed.13Fig. 11. The sensitivity to the radiative rate for Ey > 53 MeV to the variation in gp(gp'). The efficiency is folded in.A test run of ~20 h was made to observe y- rays with the TPC and to check some of the Monte-Carlo results. A ir- beam was stopped in targets of CH2 and C, and the 129 MeV y-ray peak produced in the reaction ir~p -*• yn was observed. An 0.6 mm Pb sheet was used as a converter for the lower TPC sector (#5). The lead was placed on the existing inner trigger scintillation (15) and a new veto scintillator was placed on top of the Pb. The gate width for the anode and cathode ADCs of the TPC was increased from 2 to 3 ps, but only about 50% of the fiducial volume of the TPC was used for reasons of convenience - a minimum number of hardware changes was employed in order to reduce the impact of this short test on the p e experiment. The data were analysed with a simple-minded mul­tiple track-searching routine. The subtrac­tion of the momentum histograms for CH2 and C was performed with a normalization based on the number of pion stops [Fig. 12(a)]. The results are fairly clean. The measured effi­ciency is 0.4-0.8% with all software cuts applied depending on the magnetic field. This is in fairly good agreement with Monte- Carlo runs made with the same geometry. The resolution improved at high fields and reached <10%, as shown in Fig. 12(b). Monte- Carlo studies also predict that the resolu­tion will improve and that the efficiency will be nearly doubled if inefficiency due to the reconstruction of the double tracks could be eliminated by demultiplexing.Work is now in progress in conjunction with Expts. 277 and 284 on possible demultiplexing schemes of the TPC which will permit multiple track detection. Monte-Carlo and softwareanalysis techniques are being developed to evaluate the various possibilities.Experiment 270Test o f charge symmetry by a comparison ofn~d-^nn with n +d -*p p  (B.M.K. Nefkens, UCLA)The planned technique for the calibration of our neutron detectors employs the very same reaction as the one of interest in our exper­iment namely ir-+d n(l)+n(2). One neutron in the final state, n(l), is detected in a small array of neutron counters (that can have an unspecified efficiency). n(l) defines the complete kinematics of the other neutron, n(2), creating a narrow, monochromatic, tagged neutron beam. The size and divergence of the tagged neutron beam is determined by the size and location of the tagging array as well as the size of the deuterium target, the dimensions of the pion beam, etc. A very attractive feature of this arrangement is the possibility of 'trying things out with pro­tons', e.g. setting the timing of counters and determining the actual size of the tagged beam using charged particles obtained in the kinematically identical reaction ir+d ->- p+p.We have demonstrated the value of this neu­tron detector calibration technique in a test run using channel Mil from November 18-21. The experimental layout for our calibration experiment is shown in Fig. 13. Incoming pions were defined by a coincidence between Bj and B2, two thin small beam counters and the anticoincidence with the beam halo hole counters H^2 (a double-ended counter), thusTr(in) = Bj • B2 • H j2 .The liquid deuterium target was a 3 in. dia­meter cylinder, 4 in. high with 5 mil mylar walls; the vacuum box had 5 mil mylar windows as well. The target box was surrounded on left, right, upstream and downstream sides by four anticounters A^ that covered about half of the solid angle,ir(stop) = ir(in) • A^ .We measured Tr(stop) to be 3% of ir(in).The tagging neutron counter array consisted of eight solid scintillators, made of NE110, labelled N1 to N8 or Nx . Each counter was a cylinder 19.5 cm long and 7.6 cm in diame­ter viewed by a photomultiplier. The spatial arrangement is shown in Fig. 14(a). The array14264. I 256 . i 248  ( 240 . i 232  i 224 i 216 . < 208. i 2 0 0 . i192 i 184 ' 176 ' 168 ' 160 i152. i144136.128120112104.4.5 kG ( 7 t - + C H 2 ) - U - + C )IX  2XX6XXXXXXXXXXXXXXXXX3XXXXXXXXX8xxxxxxxxxxxxxxxxxx2 XXXXXXxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx5XXXXXXXX7 0X X X X X X X X X 6xxxxxxxxxxxxxxxxxxxxxxxxXX X X X X X X X X X X X 5xxxxxxxxxxxxxx xxxxxxxxxxxxxxx X8 xxxxxxxxxxxxxxx •8XX  XXX X X X X X X X X X X X X 6  3 XXX X X X X X X X X X X X X X X X X 9  46 78X XXX X X X X X X X X X X X X X X X X X 7 X X 4 75- 1 6  I-2 4  I- 3 2  i -4 0  i101 . 00()B IN  CEN TERSFig. 12. (a) 129 MeV y - r a y  peak from the reaction ir-p Yn> (b) Pictorialview of an event.15Shielding block fo r space in beamline wallIron wallTracksBeam trap  blocksBeam pipeFRONTN ARRAYSIDENEMO Scintilla tor4 0 8  ScintillatorM ARRAYFig. 13. Experimental layout for neutron detector calibration test.was located 2 m away on the left from the target at an angle of 45 ° with respect to the pion beam. The special test neutron counter counter consisted of four contiguous scintil­lator blocks, each one 12 in. high, 2 in. wide, and 5 in. deep. Each of the blocks was viewed above and below by a 2 in. photo­multiplier via an 8 in. long air light guide painted inside with a white diffuse reflector [see Fig. 1 4 ( b ) ] .  The individual blocks were labelled M1-M4 or M y. We recorded the pulseheight of every Nx  and My counter as well as all TOFs and made a large variety of scaler readings. The high voltage of each neutron counter was set using the Compton edge of a 60Co source, corresponding to a neutron energy of 4 MeV. Each counter anode output was fed into a constant fraction dis­criminator in order to optimize the timing.Shown in Fig. 15 is the TOF spectra of coun­ter M taken with respect to B1 when a coinci­dence with N is required. It shows that the background in the coincidence measurement of tt d -*• nn is less than 2 % .We obtained simultaneously 8 calibrations of the M counter which enabled us to investigate efficiently the geometry of the cone of the tagged neutrons. The test was repeated withFig. 14. (a) Arrangement of the taggingcounters N1-N8, (b) arrangement of the test counters M1-M4.counter M located at 1, 1.25, 1.50, and1.75 m for further tests of the geometry, alignment, etc. Preliminary results are as follows: 1) the efficiency of the M counterfor 110 MeV neutrons is 11%, in excellent agreement with design calculations, 2) the efficiency of the N counters agrees with the calibrations made previously at Los Alamos.Inspection of Fig. 15 shows that the timing resolution of the neutron counters is very good. Typical values are T = 0.9 ns and full width at the base ** 3 ns . There is very little background when a coincidence is required between the two neutrons, typically less than 2%. This is due in part to the very low background in the Mil cave, the in­stallation of extra shielding and a beam dump trap. The singles spectrum of neutrals in the M counter was also quite nice. Some of the single counts originated from beam inter­actions in the beam counters. In the complete experiment we can eliminate many of these by placing the beam counters upstream and using an appropriate shielding arrangement. The only nuisance background associated with the deuterium comes from the reaction ir~d -*■ ir°nn. Since the energy of the neutrons in this re­action is less than in ir“d ■> nn, these neu­trons appear on the right side of the mono­16TOF (ns)Fig. 15. TOF spectrum for the reaction ir~d * nn measured between the incident ir“ and the first neutron where both neutrons are detected.Fig. 16. TOF between beam counter and TRIUMF pick-up coil (a) for all beam particles, and (b) for pions in beam only (beam particles that scatter into N counter).chromatic peak. We have calculated the TOF spectrum of the neutrons from ir-d > Tr°nn based on a three-body phase space with uni­form density.It hardly needs elaboration that the comple­mentary part of the experiment, namely ir+d ->■ pp, worked extremely well.The contamination of the pion beams has been studied using TOF spectra taken between the beam counter Bj and the TRIUMF proton beam pickup coil. An example is shown in Fig. 16(a). The large peak is due to the pions, the small shoulder on the right due to muons, and a little hump far to the right due to electrons. To show that these small glitches were not pions, we obtained a TOF spectrum requiring the beam particles to be scattered into the M or N counters. The result is shown in Fig. 16(b) and clearly shows that our interpretation of Fig. 16(a) is correct.Experiment 277Branching ratio o f n°-~e ‘ e (C. Waltham, TRIUMF/UBC)The rare decay ir° -»■ e+e“ is the least under­stood of the electromagnetic decays of the neutral scalar mesons. The most important decays of the it0 are as follows:Decayyy _e+e y -*■ e+e“e+e~ + e+e~Theoretical branching ratio98.82%1.184%3.2 x 10-5 ~6 x 10“8Of these only the primary decay can be con­sidered to be well understood. The Dalitz decay (ir° e+e~Y) is the subject of TRIUMF Expt. 217 (see p. 10).Experimental measurements of the branching ratio B = T (ir0+e+e-)/r (tt°->TY) are much higher than expected from plausible theoretical mod­els, which predict values only slightly high­er than the QED unitarity lower limit of4.75 x 10-8. Current published measurements are those of Fischer et al. [Phys. Lett. 73B, 359 (1978)] at CERN using tagged ir°s fromK+ -*■ TT+ir° decays and Mischke et al. [Phys. Rev. Lett. 48, 1153 (1982)] at LAMPF using ir°s from the charge exchange reaction Tr“p -*■ 7T°n in flight. Their values are (2.23±^]jJ x 10_/ and (1.8±0.6) x 10_/, respectively.The unitarity lower bound is model independ­ent and arises from the imaginary part of the amplitude for a 2y intermediate state. B may be increased by introducing a real part which requires the xyy vertex function to couple with off-shell photons. The result is loga­rithmically divergent, but may be made finite with specific model assumptions. Effects of17vector dominance and quark loops (related by Q2-duality), weak neutral currents and mas­sive Higgs bosons also contribute to this process. The only effect which may decrease B is that of CP-violatlng neutral currents.(a) Detection of the 420 keV neutron from ir-p ir°n and measuring its energy to <1% FWHM by time of flight. This is a method of tagging real ir° production and reduces only the in­ternal conversion background.The most recent calculations use the quark- loop model [e.g. Pich and Bernabeu, Z. Phys. C22, 197 (1984)] and all yield a value of~6xl0-8 for a wide range of quark masses. Ef­fects of weak neutral currents, Higgs bosons and CP violation are expected to be small. The only remaining way to increase B signifi­cantly is to invoke anomalous lepton-hadron couplings, but one is reluctant to do this, as values of up to 2.3 x 10“ 3 are then easily obtained! Clearly the situation calls for clarification.The TPC at TRIUMF is a natural detector to use to observe tt° •* e"*"e-. It has a large ac­ceptance and efficiency and, after the read­out electronics have been demultiplexed, it will be able to detect the double tracks of the e+e“ pairs. A copious flux of tt° s  can be produced by stopping a beam of tt~ in a liquid hydrogen target inside the TPC. The back­grounds are very large, however, and are due to the following processes:1)2 )  TTC3) TTC4) ircTT p ne+eY+X +*■ e+e“Ye+e“+Ye+e e+e“(internal conversion) (external conversion) (Dalitz decay)(double Dalitz decay)Monte-Carlo calculations show that with the ~6% momentum resolution that the TPC current­ly has the it0 -*■ e"*"e_ signal would be com­pletely swamped by these processes. Two ap­proaches are currently being evaluated to reduce the backgrounds:(b) Optimizing the momentum resolution of the TPC. This reduces all the backgrounds beneath the signal, but the resolution needs to be better than 2% FWHM in order for the signal to be seen unambiguously (i.e. a visible peak in the e+e“ invariant mass spectrum).Two different prototype neutron detectors have been constructed, and in November they were shipped to the Van de Graaff laboratory at Queen's University for testing in a low energy neutron beam. Signals from 420 keV neutrons were clearly visible, but problems with data acquisition made it difficult to extract the timing resolution, and it was not clear that this was adequate for the task (<2.5 ns FWHM). Data analysis is proceeding.Much more encouraging were tests on the TPC in October aimed at optimizing the momentum resolution. The magnetic field was increased to its maximum of 9 kG and the trigger was changed so that the monoenergetic positrons from ir+ * e+v (69 MeV) could spiral inside the chamber. This improved the resolution from ~6% to 3.0% at this energy.Further tests are planned for 1985 to opti­mize the- resolution by minimizing the mass encountered by the electrons and by improving track-fitting techniques. If a resolution of 1% can be achieved, then the branching ratio B can be measured to ±10% (if B ~ 6*10-8) in a month of continuous beam time. A resolution of 2% would require four times as long.18NUCLEAR PHYSICS AND CHEMISTRYExperiment 142Complex reaction mechanisms(R. Korteling, Simon Fraser)Experiment 142 is designed to study the ques­tion whether complex reactions initiated by protons are dominated by relatively few di­rect interactions or whether they proceed primarily through some intermediate statisti­cal source. It is configured to measure the correlation between an emitted light frag­ment (p,d,t,3He,4He) and a relatively high energy forward-going proton, believed to be the incident particle. An initial four light- fragment telescope systems are located at fixed angles of -60°, -90°, -120° and 90° to the beam with respect to a movable proton telescope (10° < 0 < 70°). It is intended to add two light-fragment telescope systems at -30° and -150° before the final data are collected.An initial test run in June indicated that, except for a halo problem at the BL4B TI lo­cation, the experiment could collect data as then configured. The halo problem is a con­sequence of a very large circulating beam in the cyclotron (>100 yA) with a requirement of a low current (5-20 nA) beam for this experi­ment . TRIUMF is currently working on ways to suppress the background. In the meantime the count rate capability of the front proton detector, which was the limiting system, has been improved by changing from one based on Nal to one based on a plastic scintillator. A recent run in November indicated that the background would probably be handled with the new detector system, with some shielding. In addition the TRIUMF solution for the problem is expected to be installed in the near future (before final data collection).Preliminary results from the November run indicate that the Z=2 energy spectra are very similar from Be and Ag targets, both in sin­gles and coincidence with the forward proton. However, the coincident proton angular depen­dence is quite different. The results from Be seem to show the same strong dependence on the QTBS angle observed for the simpler (p,2p) and (p,pd) measurements. The Ag re­sults seem to be less pronounced, with the possibility that the peaks are smoothed out. In both cases, however, the fragment spectra are shifted in the coincidence mode relative to the singles consistent with the selection of events with a perpendicular component inthe coincidence case. In general, the results indicate the importance of obtaining as com­plete a kinematic description as possible before drawing any conclusions about these complex reactions.Experiment 169p + 160 elastic scattering(D. Hutcheon, Alberta/TRIUMF)The final 'clean-up' run for this experiment, which was twice postponed due to delays in the MRS upgrade program, took place in June. The p+*80 elastic scattering cross section was remeasured at intermediate angles (16° to 45°) at 200, 300, 400 and 508 MeV. The high- resolution dispersed beam mode of MRS opera­tion was used at 200 MeV to disentangle the p+^Ca elastic scattering from scattering due to lighter contaminant nuclei in the 15° to 20° region. These data have been analysed and, combined with our previous measurements, will be compared with current relativistic and nonrelativistic models of elastic scat­tering .Experiments 173, 196, 204, 231, 250Broad pionic X-rays (A. Olin, Victoria/TRIUMF;G. Beer, Victoria; J. Bailey, DESY)The broadest pionic X-ray lines observed so far have had anomalously narrow widths, and attempts to describe this phenomenon within the framework of the optical model have not been successful. Our program aims at remea­suring some of these lines (*8>180, 22Ne,23Na 2p-ls and 208Bi 4f-3d) and extending the range of these measurements to other nuclei (24Mg, 27A£ 2p-ls and 208Pb 4f-3d). Theselines are very weak as well as broadened, so a segmented BG0 Compton suppression spectrom­eter [Olin et al., Nucl. Instrum. Meth. 222, 463 (1984)] has been developed to reduce the backgrounds. An overall factor of 4 reduction in the photon background was achieved in beam. Data-taking on all these elements has now been completed. The X-rays from Mg, A I  and Pb were clearly observed for the firsttime. In an attempt to further extend the range of these measurements we have also looked for X-rays from 31P in coincidencewith pion single charge exchange and with energetic neutrons. It was hoped that this technique would lead to a strong suppression of the Y~ray backgrounds from pion absorption.19Pb r„'‘MiiMliM!0.2  0.3 0.4 0.5A  FM0.2 0 .3  0 .4  0.5A  FM0.2  0 .3  0 .4  0 .5A  FM* 0 .0  0.1 0 .2  0 .3  0 .4  0.5A  FMFig. 17. Level shifts and widths are shown as a function of the difference of the neutron and proton matter distribution rms radii. The cross-hatched band shows the experimental value and its lo error. The calculations use potentials from Kunselman et al. [Nucl. Phys. A405, 627 (1983)] A , Batty et al. [Nucl.Phys. A322, 445 (1979) • and Seki (private communi­cation) ■ .The analysis of the Pb and Bi data has now been completed, and the results are given inTable III. The Bi 4f-3d energy and width dis­agree strongly with the earlier measurement. In Fig. 17 the strong interaction shifts and widths for the 3d and 4f levels are compared to optical model calculations. The width ofthe 2 0 8 Pb 4f-3d transition is about 30% low, which may be an indication of the saturation phenomenon calculated recently by Tauscher et al. [Nucl. Phys. A415, 333 (1984)].The Na data accumulated in 1983 contained a similar statistical sample to that of our 1978 work. However, due to the use of the Compton suppressor the background under the peak has been reduced by a factor of 4, and the Compton edge that lies under the peak was also greatly suppressed through the use of a thin, low density target. Fitting the data over the same region and with the same y - r a y  contaminants used in the analysis of the 1978 data produced a comparable width of 13.4 keV. However, as can be seen in Fig. 18, the back­ground obtained from this fit does not ex­trapolate very well to lower energies. Close examination of the region near the top of the peak (Fig. 18) showed a hint of two addition­al gammas in both data sets. This complexstructure makes Na a particularly unfavour­able case for a precise measurement. Extend­ing the fit region and including the possi­bility of these additional gammas gave a final width of 16.7±3.1 keV and an energy of 276.2±0.9 keV. These two factors produce the dominant contribution to the systematic un­certainty of ±2 keV in the width and±0.65 keV in the energy. So in this case toothe evidence for narrow widths is weakened.The 2p-ls X-ray is seen very clearly in our Mg data, and contaminant gammas are not a serious problem. In AH the background and contaminant problems would appear to be some­what worse than the Bi 4f-3d X-ray, and large systematic uncertainties are expected.The isovector s-wave optical parameters have been determined mainly by the 160 and *®0shifts and widths. We have taken ComptonTable III. Measured pionic X-rays in Pb and BiTransition Energy error EM(keV stat sys) (keV)Shift Width(keV) (keV stat sys)AreaPb 4-3 1274.7+0.6±1.0 1251.0 23.7 47.4±2.6±4.6 16,200Pb 5-4 575.44±0.02±0.05 573.80 1.64 1.30±0.03±0.08 270,000Bi 4-3 1312.0±1.8±3.0 1282.40 30.4 71.2±8.4±21 15,000Bi 5-4 589.90±0.05±0.04 588.13 1.77 1.49±0.02±0.04 284,0002 .5 0 0 r  B i  ^2 .3 0 00 .0  0.1 0.2 0 .3  0.4A  FM0.2 0.3A  FM1.5000 .7 0 020EN E R G Y  (K E V )Fig. 18. Photon spectrum of 23Na In the region of the 2p-ls X-ray. The extrapolated fit to region #1 (255-293 keV) is shown together with a fit to a wider region. Inset: expanded plot showing possible contaminations on the top of the X-ray peak.suppressed data on these isotopes and, more Experiment 189recently, on isotopically pure 22Ne. This Radiochemical study o f Oj{E) for 209Bi (p,n f 10 xAtwork complements the low energy pion scatter- from threshold to 800 M eVing measurements on these targets. (J. D ’Auria, Simon Fraser)In a test run this summer we used the CSS andtwo large Nal(TA) detectors to search forX-rays from 3*P in coincidence with pion single charge exchange (CEX) and with ener­getic neutrons. The CEX rates were rather high, as expected from the favourable selec­tion rules and Q values, but appeared to oc­cur mainly from the pionic 3d level. A pre­liminary evaluation of the data indicates that the technique will not be useful in studying broadened X-rays, but the CEX cross sections will be of considerable interest.We have now constructed a xenon gas scintil­lation proportional counter in preparation for another attempt at the 2p-ls transition in D2 gas. It is expected that with a gascounter we will be able to remove the back­grounds that plagued our earlier work with Si(Li) detectors. This device is expected to be tested in early 1985.The energy dependence of the total angle- integrated cross sections for the series of reactions 209Bi(p,Tr-xn)21®-xAt, with x = 0-7, has been studied from below threshold to 800 MeV at IUCF, TRIUMF and LAMPF using acti­vation and radiochemical techniques. Residual nuclei with 5-7 neutrons removed from the coherent product were found to be favoured in the (p,Tr~xn) reaction channel. The excitation functions (Fig. 19) show an energy dependence similar to that of the previously measured 209Bi(p,ir°)2^Po reaction [Ward et al., Phys. Rev. C 24, 588 (1981)]. Residual asta­tine mass distributions are consistent with a Gaussian shape that implies an average resid­ual excitation energy of ~60 MeV and average momentum transfer of ~335 MeV/c for the (p,7r“xn) channel. At incident energies below ~250 MeV the (p,ir-xn) channel dominates ir- production. Above 250 MeV the increasing probability of charged particle emission keeps the total (p,ir_xn) cross section at a relatively constant value accounting for21100101toa .10-1:101-i  203A t  . V**i •204A t208A t.Qa .101 lT  207A tf • • •  •209At? i •1208A t210 A tI * t t  T T tT1 1 1 i 1 r-0 200 400 600 800 200 400 600 800E (MeV)p •E (MeV)PFig. 19. Excitation functions for the yield of At nuclides from the series of (p,ir“xn) reactions on 209Bi. The open squares are the results of Clark et al. [Phys. Rev. C 26, 2073 (1982)]. The dashed curves are the pre­dictions of Gibbs [Pion Production & Absorp­tion in Nuclei, AIPCP #79 (AIP, New York, 1981), p. 297]. The 210At graph also displays the excitation function for 209Bi(p,ir°)2l0Po as a solid curve.<0.5% of the inclusive (p,ir ) cross section(Fig- 20).Fig. 20. Excitation curve for the summed (p,Tr“xn) cross sections. The open square is from Clark et al. (op. cit.). The solid and dashed curves are predictions of Gibbs and Long et al. [Phys. Rev. C 26_, 586 (1982)],respectively. The open circle and triangle are the experimental results of Crawford et al. [Phys. Rev. C 22_, 1184 (1980)] andCochran et al. [Phys. Rev. D JLl^  3085 (1972)]respectively. The crosses are recent unpub­lished results of DiGiacomo et al. [LAMPF preprint LA-UR-84-1553]. The closed tri­angles are scaled data from Krasnov et al., Phys. Lett. 100B, 11 (1982)].Experiment 199Coincidence studies o f n- absorption on 3Heat 85  M eV  (A. Altman, UBC/TRIUMF)Pion absorption on 9He has been studied by measuring the proton-proton coincidences coming from the absorption of n+ on 9He. We have reported on the differential cross sec­tions for ir+ absorption at 65 MeV pion bom­barding energy [Moinester et al., Phys. Rev. Lett. 52, 1203 (1984)]. For the new measure­ments we constructed a thicker and cylindri- cally symmetrical target. With this new target we measured the target thickness via elastic scattering and range techniques. We find good agreement for the 65 MeV data taken with the old target compared to the new target if we renormalize the earlier results [Moinester et al., op. cit.] by a factor of 0.83. In Fig. 21 we compare our current measurements with the previous results.The angular distributions in Fig. 21 are presented in the ird centre of mass. The shapes of the angular distributions are well described by the ir"N-d->-p+p angular distribu­tion [Ritchie et al., Phys. Rev. C 27_, 1685(1983)]. The ratio of the two-body absorp­tion cross section of it'1" on the T=0, S=1 pair in 9He compared to n'*’+d->-p+p at 65 and 85 MeV is 1.38±0.08 and 1.50±0.08, respectively.The angular correlation of coincident nucle­ons is also of interest because it reflects the range of internal momenta in % e  which are important for the two-body absorption mechanism. In our measurements we see a full width at half maximum of 12.4°±1.0°. This corresponds roughly to a spectator momentum of less than 60 MeV/c. The narrowness of the angular correlation shows that the ir+ is preferentially absorbed on two nucleons whose total momentum is also small. Calculations2260 a  ioo u cmFig. 21. Angular distribution of the 3He(it+ ,pp) reaction at 65 and 85 MeV. Crosses are data from Mo inester et al. multiplied by 0.83. Closed circles are new Lee and Otha [Phys. Rev. Lett. 49, 1079(1982)] which include large contributions from an intermediate A in the absorption process are able to account for the narrow­ness of the angular correlation at 165 MeV bombarding energy.Pion absorption on a T=1 3Sq nucleon pair was studied at 1^=85 MeV at TRIUMF by detecting a p*n coincidence from the 3He(ir-,pn)n reac­tion. Unlike the absorption on T=0 3S^ pairs, for the *Sq pairs intermediate A-N states cannot have LAN=0. Therefore, this cross section may be suppressed, and there is an increased sensitivity to other processes. The three-body part of the absorption process follows the three-body phase space and was subtracted from the spectra to obtain the two-body contribution. In the energy spectra of the detected nucleons (Fig. 22) we observe a peak at the energy corresponding to two- nucleon absorption. The other peak in the spectrum corresponds to the final-state int­eraction case where the two outgoing neutrons follow the same direction as if they were a single particle. The angular distributions for the (ir~,pn)n reaction (Fig. 23) show a significant asymmetry about 90° c.m., which is a signature of a mixture of even and odd partial waves through non-A absorption. The angular distribution data were fitted to Legendre polynomial in the form:da/dft = 53 AiPi (cos6) (Fig. 23).Fig. 22. Energy spectrum of protons from the 3He(ir-,pn) reaction at 85 MeV.The ratio of the two-nucleon cross sections between (ir+,2p) and (ir~,pn) reactions is R = 15.3±1.4 at Tn = 65 MeV, R = 15.0±0.9 atRev. Heidel-T-n = 85 MeV [Moinester et al., PhysLett. 52_, 1203 (1984); Aniol et al. berg contribution].The independent determination of the ^He target thickness enabled us to observe the ir-+3He n+d reaction. In Fig. 24 we display the differential cross sections of this reac­tion as measured by this work. We also show results derived from other measurements.Fig. 23. Angular distribution of the 3He(ir~,pn) reaction at 65 and 85 MeV. Crosses are data from Moinester et al. multiplied by 0.83. Closed circles are new data.2380 100 120 140Fig. 24. Differential cross section of the x-+ 3He n+d reaction. The symbols are •65 MeV, ■ 85 MeV this experiment; * T^- (equivalent) = 129.5 MeV, o T^- (equivalent) = 66.1 MeV, O T^- (equivalent) = 82.7 MeV from p+d ■> t+ir+ of Lolos et al. [Nucl. Phys. A238, 477 (1982)], Silverman, Ph.D. thesis, UCLA (1982)]; A T^- (equivalent) = 67.5 MeV from p+d + ir°+3He of Lolos et al.; Silver­man [op. cit.]; x T^- = 50 MeV, V T^- = 100 MeV from ir"+3He + b+d of Kallne et al. [Phys. Rev. C 24, 1102 (1981)].Isospin dependence of pion absorption by 12C at = 65 MeVThe (ir+,p»p) and (ir~,p«n) reactions on 12C were measured with 65 MeV pions on the Mil channel at TRIUMF. The contribution of the two-body absorption of the it to the total absorption cross section can be estimated and the ratio for absorption on p-n pairs to that on like nucleons can be deduced. The two nuc­leons were measured in coincidence. Figures 25 and 26 show the angular correlation for ir+ and tt“ with one proton detected at 65°. The distribution was fitted with a Gaussian and a three-body phase space shape for the back­ground (absorption on three nucleons in *2C). The fit was used to integrate over the corre­lation. The results are summarized in Table IV.The ir+ two-body (quasideuteron) absorption together with the scaled (Deff) ir+d>p*p reac­tion cross sections are shown in Fig. 27. From the normalization factor Deff of the two-body absorption to the ir+d+p'p process0I____I___L___ I___ I_i - L______ I___ i--- 1-50  60  70  80  90  100 110 120 130 140 150Qp (deg)2Fig. 25. Angular correlation for 12C(ir+ ,p,p). The arrow is at the kinematical position for x+d->-p*p. The curves are the fits to the data (see text).9n ( deg )Fig. 26. Same as Fig. 25 for l2C(ir“,p*n).Table IV. The reaction.T»(MeV)a2N(n+ .p -P)(mb)02N (u_ >P*n )(mb)1T+aabs(mb)Def f65 24.216.0 2.84H.1 97±23b 3.2710.8165a 17.812.7 194t36b 1.4710.1245a 11.412.0 95i32b 2.7010.2aFrom Altman et al., Phys. Rev. Lett. 50, 1187 (1983) ^From Navon et al., Phys. Rev. C 22, 717 (1980); Phys. R ev. C 28, 2548 (1983)24Fig. 27. Angular distribution of the quasi- deuteron component of 12C(ir+ ,p»p) in the c.m. system of ir+d+p'p. The curves are the ir+d->- p*p cross section normalized to the data.and the total cross section [Ritchie et al., Phys. Rev. C 2h_, 522 (1981)] of the latter we get the integrated two-body cross section 02N (*+ ,P-P). With Table IV we see that the effective number of deuterons Deff is small­est at resonance energy where the mean free path for the it is shortest. The ratio of the two-body cross section 02N(7r+»PP) t0 t^le t0_ tal cross section indicates a two-bodycontribution of at least (25±9)%. The true percentage is higher due to final-state interaction losses. The ratio 02n (tt+ ,p *p )/ a2N(lr_>P*n) (8.5±2.6).Experiment 206A study of(p,n) and related reactions (J. D'Auria, Simon Fraser; R. Heimer,Western Ontario)Data-taking continued in the early part of the year for this experiment, whose primary purpose is to compare inclusive (p,n) and (p,p') reaction yields at several angles for a few targets at 190 MeV and 500 MeV.Last year we reported on our success in em­pirically arriving at a combination of iron shielding around the neutron detectors and p-metal shielding around their photomultipli­er tubes that allowed us to move the detec­tors from one angular location to another without the accompaniment of intolerable gain shifts in the tubes. This year the majorstumbling blocks were the presence of a large background of scattered particles which caused a very poor signal-to-noise ratio in the detectors, particularly in the Nal detec­tors used to detect the protons, and the dif­ficulty that we experienced in extracting a beam of low enough intensity down beam line 4 that dead times in our counting system were not excessive.The background is attributed to beam which is singly stripped, either electromagnetical- ly or via scattering off residual gas, while still in the cyclotron. This now neutral component of the beam leaves the site in a direction tangential to its original orbit, and those particles which happen to be directed towards the vault section of beam line 4 travel down that line until they are stripped again. Through some poorly under­stood mechanism many of these protons are bent or scattered into the beam line 4B area. The orbit in the cyclotron which is tangen­tial to beam line 4 is at a radius corre­sponding approximately to 400 MeV, and indeed the background problem was worst when the beam line was tuned to this energy. This 'beam' had been observed previously, but had not caused difficulty in those earlier exper­iments because they involved coincidence measurements which required much higher beam currents than the present singles experiment can tolerate without suffering excessive dead times.The other difficulty experienced is in achieving large split ratios for low current running in beam line 4. It seems that as the current delivered to beam line 1 has been in­creased over the years, so has the circulat­ing beam risen vertically in the cyclotron. This has made it more and more difficult to get stripper 4 out of the beam. Our solution to this problem was to run at 190 MeV oppo­site 30 pA in beam line 1, so that the needed split ratio was not so large, and to run at 500 MeV, where stripper 4 could be shadowed by stripper 1, rather than at 400 MeV as originally planned.Both procedures help to solve the background problem because the background scales with the beam circulating in the cyclotron (this improved the situation at 190 MeV), and the background at 500 MeV is at a tolerable level.Data were collected at five angles (24°, 59°, 83°, 117°, 143°) for three targets (Be, Ag, Ta) and two energies (190 MeV and 500 MeV).25B E A MMAGNETVETOSLEAD "GLASSPHOTONDETECTORS,^ " shielding-GASEOUS # 2  -LIQUID H2 -S -R A Y  MAGNETVERTICAL DRIFT CHAMBERSPLASTIC SCINTILLATORSVERTICALDRIFTCHAMBERSSCINTILLATOR TELESCOPE AND VETOAFig. 28. Experimental layout.The analyses of the proton spectra are essen­tially complete, and a start has been made on analysing the neutron spectra. The latter must proceed in several steps, the first of which (obtaining the drifts in the reference times for the TOF spectra) has been completed.POSITION IN TARG ET (cm )Fig. 29. Triple coincidence (ppY) rate as a function of origin of event in target.problem now seems to be under control. The experimental layout is shown in Fig. 28.In Fig. 29 is shown the number of events de­tected as a function of position in the tar­get, as reconstructed from the event co-ordi­nates in the front two drift chambers. The two histograms show the distribution for equal amounts of incident beam for the cases when a) the target cell is full of liquid hydrogen and b) when the target cell contains only residual hydrogen gas. This figure demonstrates that the events we believe to be ppY events do come from hydrogen, and not from the target walls, etc. Figure 30 shows the energy distribution of high energy pro­tons detected in the magnetic spectrometerExperiment 208  Proton-proton bremsstrahlung (P Kitching, Alberta/TRIUMF)During 1984 the apparatus for Expt. 208, ameasurement of proton-proton bremsstrahlung with polarized protons, was installed and commissioned. Major milestones were the com­pletion and installation of the scattering chamber and liquid hydrogen target in March, of the spectrometer bending magnet in August and of the four vertical drift chambers in October. Four commissioning runs took place. The most serious problems encountered were due to poor beam quality on beam line IB, which led to unacceptable accidental coinci­dence rates swamping the real ppY events. After improving the machine tune, installing two collimators in the beam line, and making a substantial effort to pinpoint the sources of background and improve shielding, thisO I - pc° IHIGH ENERGYPROTON I I 1 * •ELASTIC ALLY SCATTERED PROTONStgjprrfMfHtr t • •isu$*. a. « n n 4Fig. 30. Kinematic locus of events believed to be from ppy reaction.26for five values of the angle of the corre­sponding low energy proton. The energy scale was calibrated using the elastically scat­tered protons shown at the top of the figure. Also shown are the kinematic limits on the energy of the high energy proton for the con­ditions shown. It can be seen that the events believed to be from ppy interactions do obey ppy kinematics.We are now at the stage where we believe that we are ready to take data. This will be done in the next two polarized beam periods in January and May of 1985.Experiment 212In search o f a tredecabaryon resonance(K.P Jackson, SFU/TRIUMF)The beam time allocated to Expt. 212 waslargely devoted to commissioning the MRS up­grade with particular emphasis on measure­ments of small cross sections at large momen­tum transfer. This choice was dictated in part by problems apparently associated with the operation of beam line 4B concurrently with beam line 1A at 130 pA and in part by the fact that this was the first experiment to utilize the new MRS hardware with disper­sion matching of the beam line to the spec­trometer and in measurements at large momen­tum transfer. Spectra were recorded of the elastic and inelastic scattering of protons incident at 362 MeV on targets of carbon andch2.Initially an achromatic beam was used todetermine the variation in the acceptance ofthe spectrometer as a function of both the vertical position of the beam on target and the deviation of the momentum of the scat­tered proton from that of the central trajec­tory in the spectrometer. These measurements, conducted at a scattering angle of 25.4°,were followed by similar measurements with the beam line tuned to achieve momentum dis­persion of -10 cm/% at 4BT2. This tune re­vealed the existence of two rather well de­fined components in the beam separated inmomentum by ~0.2%. The total observed energy spread (~2 MeV) was significantly larger than during previous runs for Expt. 212. The mea­surements with this dispersed beam were also used to investigate the problem of the norma­lization of cross sections associated with the transport of the large beam envelope through the beam line.Additional measurements were made at scatter­ing angles from 42.5° to 92.0°. For the larg­er angles an additional plastic scintillator mounted in the region of the focal plane was used to improve the rejection of background events resulting from random coincidences. Data were recorded under conditions for which the count rate in the wire chamber located at the front end of the MRS was as high as 0.8 MHz and with beam currents as high as 100 nA incident on a 0.1 g/cm2 carbon target. The analysis of these data is in progress and will be used to critically assess the per­formance of the upgraded spectrometer in the measurement of small (<1 nb/sr) cross sections at very large momentum transfer (>1 GeV/c).It is apparent that the successful completion of this experiment will depend critically on the highly efficient tuning of beam line 4B at all energies, on the quality of the beam delivered to 4BT2 while operating beam line 1A at maximum current and on the optimum per­formance of the spectrometer. Requests for additional shifts for Expt. 212 will be deferred until these issues are resolved.Experiment 213Negative pion absorption a t rest in light nuclei(N. Grion, Trieste)The aim of this experiment was to a) measure and compare the inclusive neutron, proton and deuteron emission after stopped it" in light nuclei (6Li, 9Be, *2C, and 2 / A l )  abovethe evaporative region (E > 20 MeV), andb) measure the (n,n), (n,p) and (n,d) corre­lations following pion absorption at rest in 9He .Exclusive 9He(ir-,nx)x, x = n, p or d measure­ments will provide knowledge on the multinuc­leon role in the absorption process, that is, on the contribution of a genuine four-body cluster to the absorption process versus the quasi-free two-nucleon absorption and, in this latter case, on the influence of the pair of spectator nucleons on the two in­volved in the interaction. This understand­ing could be used as input for calculations on the pion absorption process in more com­plex nuclei where on the average more than two nucleons are involved. Inclusive data will provide information on the dynamics of the pion absorption mechanism in complex nuclei, namely, on the effects on the N-nucleus final-state interactions and the number of nucleons participating in the ab­sorption process. The strong isospin depen­dence of the ttNN absorption process will be27described by the ratioprobability (ir~np nn) np probability (ir-pp *  np)which is a quantity measured in the present experiment.Inclusive data have been reduced to their final form, while 9He data are being analysed Carbon data have already been published [Cernigoi et al., Nucl. Phys. A411, 382(1983)].Figure 31 illustrates the inclusive neutron energy spectrum (dots) of the 160(ir-,xn) re­action. It is compared with theoretical pre­dictions of Jackson and Brenner [Prog. Part. Nucl. Phys. _5, 143 (1982)] (double dashed-double dotted curve) and Chiang and Hufner [Nucl. Phys. A352, 442 (1981)] (dashed dotted curve). In Jackson and Brenner the pion cap­ture is assumed to take place on an alpha- substructure of *60. In Chiang and Hufner the absorption process involves two nucleons (primary nucleons). Above 80 MeV our data can distinguish between the two proposed absorp­tion mechanisms. The two-nucleon process ac­counts for intensities in close agreement to our data. Moreover, it decreases with the neutron energy as predicted by a three-body phase space (full curve). Calculations based on four-nucleon absorption fall off earlier than our data since the kinematics of the 1+He(ir_,n) 3H reaction do not allow neutrons to exceed 90 MeV.Neutrons from the 160(ir-,n) 1SNg>s _ reaction contribute to the spectrum beyond the three- body phase space kinematical limit. The bell curve in Fig. 31 represents their intensity as a function of the neutron energy. These neutrons account for a multiplicity of (1.25±0.27) x 10-2 neutrons per stopped pion. Calculations predicting the single-neutron emission on the basis of a quasi-free two- nucleon absorption mode [Coupat et al., Nucl. Phys. A403, 497 (1983)] yield a probability (per stopped pion) of a factor of 5 lower than our experimental value. This large dis­crepancy leads to the conclusion that the basic mechanism in the single-neutron emis­sion process mainly involves one uncorrelated nucleon, namely, ir-p -*■ n.Figure 32 illustrates the inclusive proton and deuteron energy spectra after negative pion absorption at rest in The protonand deuteron spectra are compared with theo­retical predictions of Chiang and Hufner andof Datar and Jain [Phys. Rev. C 26, 616(1982)], respectively. Both of these calcu­lations use a two-nucleon vertex as a first step in describing the absorption process. In the model of Datar and Jain deuterons are formed on the surface of the nucleus by nuc­leons (mainly neutrons) that, on their way out, pick up a second nucleon (secondary deu­terons). Calculations based on this model agree well with our data, thus indicating that most of the observed deuterons are sec­ondary ones.Neutron, proton and deuteron yields (Yx) were obtained for all the studied nuclei except for 27A£, where charged particle spectra were not measured. In the region where both neu­trons and protons are primary (Chiang and Hufner) and all events come from two-nucleon absorption (En,p > 96 MeV), the ratio Rnp> of the relative importance of nn to np emission, can be evaluated as (Chiang and Hufner)Rnp = l/2x(ReXp-l) , where Rexp = ^n/Yp.The ratio R between the matrix elements for the quasi-free pion absorption by an (np) or (pp) pair is given by(ir“np -*■ nn) Z-l R = --------- - = -—  x R(iT-pp -*■ np) 2N Pwhere 2N/(Z-1) = Rstat as t*le statistical ra­tio expressing the relative probability to find an (np) or (pp) pair in the nucleus. Rexp> Rnp> Rstat and R are tabulated in Table V. Rnp values can be compared with predictions of a microscopic calculation [Shimizu and Faesller, Nucl. Phys. A333, 495 (1980); ibid., A306, 311 (1978)] for the qua­si-free two-nucleon absorption process. For p-wave absorption these calculations predictTable V. Experimental (nn) to (np) ratiosafter n nuclei.absorption at rest in the studiedRatio 6Li 9Be 160E-exp 18.0(3.0) 25.4(4.3) 15.5(2.5)Rnp 8.5(1.6) 12.2(2.3) 7.3(1.3)Rstat 3.0 3.3 2.3R 2.8(0.6) 3.7(0.7) 3.2(0.5)28NEUTRON ENERGY (MeV)Fig. 31. Inclusive energy spectrum of neutrons emitted after stopped negative pion absorption in 160 (dots).KINETIC ENERGY (MeV)Fig. 32. Inclusive energy spectra of protons (dots) and deu- terons (squares) emitted after stopped negative pion absorp­tion in 1°0.2918-E x (M eV )Fig. 33. Pion spectrum taken at 354 MeV for a pion angle of 45° lab. The energy resolu­tion is ~300 keV.a value for Rnp = 1  if the pion is absorbed directly by one of the correlated nucleon pair, and Rnp ~ 9 if the pion is rescattered following the excitation of a A resonance in the intermediate state. Our data clearly ind­icate the necessity to include the rescatter­ing mechanism in the description of the ab­sorption process if the nn-to-np ratio needs to be accounted.The ratio R seems to be independent (within the error bars) of the studied nucleus. This is in qualitative accordance with the calcu­lations of Chiang and Hufner of the ratio of the matrix elements, under the assumption that neutrons and protons are homogeneously distributed over the nuclear volume.Experiment 218Pion production from 12C  and 10B with polarizedprotons o f 350  M eV(G .J. Lolos, Regina)In November Expt. 218 received 12 shifts of unpolarized beam against 140 pA of beam ex­tracted for meson hall users. From past ex­perience such a combination has proven to be very difficult+ in achieving optimum condi­tions for (p,it-) experiments on the MRS; the November run, however, was successful in ob­taining a 12 cm/% dispersed tune with suffi­cient resolution to obtain meaningful data. An example of the resolution obtained is shown in Fig. 33; the ground and 9.50 MeV states of 13C are easily identified. Data were obtained for MRS angles 20°, 30°, 45°, 60°, 75°, 90°, 100° and 110° at incidentproton energy 354 MeV.The results are currently being analysed; preliminary indications point to a monotoni- cally decreasing and featureless angular distribution of the cross sections with the slope for the 9.50 MeV 13C* state appearing much steeper at 354 MeV than previous mea­surements at 250 MeV indicated [Lolos et al., Phys. Rev. C 3£, 574 (1984)]. The results, especially those for the stretched neutron 2p-lh 9.50 MeV 13C* state, will provide an important test of microscopic TNM calcula­tions currently in development [Iqbal, pri­vate communication].Experiment 221Search for evidence o f delta-nucleus interaction intermediate state in proton elastic scattering(H.O. Meyer, IUCF)It has been suggested that the discrepancy between data for large angle proton-nucleus elastic scattering at 200 MeV and standardoptical model interpretations can be ex­plained by the formation of intermediate A isobars [von Geramb, in Proc. Workshop on the Interaction Between Medium Energy Nucleons in Nuclei, ed. H.O. Meyer, AIPCP #97 (AIP, New York, 1983) p. 44]. One expects that the con­tribution of an intermediate A isobar in­creases when the bombarding energy is in­creased to values around 300 MeV at which the A is excited in resonance. Therefore, the A hypothesis can be put to a test by an experi­ment at this energy. To this aim we have performed measurements of the differential cross section of *2C(p,p)12C at Tp = 300 MeV for large momentum transfers. The experiment covered laboratory angles from 16° to 120° and was performed with an unpolarized proton beam from the TRIUMF cyclotron, using the MRS spectrograph in its standard configuration. The results are shown in Fig. 34.The data of the present experiment have been fitted in the framework of the conventional optical model. At this energy the real cen­tral potential is expected to deviate from the commonly used Woods-Saxon form. The shape is better parametrized by the sum of two terms, one attractive and one repulsive. However, since this form leads to a sizable parameter ambiguity because of cancellations between the two terms, we have first used a single Woods-Saxon term and determined its parameters by a fit to the forward angle data (®c.m. < 70°). Then a very short-range repul­sive term was added to improve the back angle fit, allowing only for small variations in the other previously determined parameters.30The resulting angular distribution is shown as a solid line in Fig. 34. As can be seen,the agreement with the data is quite goodover the whole angular range.Compared to the situation at, e.g., 200 MeV, there is clearly little need to invoke an ad­ditional mechanism to account for the large- angle data. Since the present data have been measured on resonance with respect to the in­termediate A isobar we must conclude that the mechanism responsible for the large-angle discrepancy at the lower energies is not the formation of a A isobar.To demonstrate this more clearly we have alsocarried out coupled-channels calculations of the type described in the reference above. In addition to the elastic channel a second channel is taken into account where the pro­jectile is transformed into a A isobar and. oEEoc?1■othe target nucleus is excited to the isovec­tor 1"*" state at 15.11 MeV. Using parameters for the coupling potential which qualitative­ly explain large-angle scattering at 200 MeV, the calculation was carried out at 300 MeV. The results is shown as a dashed curve in Fig. 34. It is evident that this model over­estimates the data by more than two orders of magnitude at the largest angles. This in­creased effect is of course due to the fact that the incident energy matches more closely the resonance energy for intermediate A states. The intrinsic energy dependence of the coupling potential is assumed to be small since the energy dependence of the elementary NN -*■ NA transition potential is small. So, we conclude that the present data rule out the possibility that the back-angle anomaly found between 100 and 200 MeV is due to the formation of intermediate A states.A brief report on this unfortunately negative but important result has been submitted for publication. It leaves us without a physical explanation of the anomalous behaviour of large-angle elastic scattering data between 100 MeV and 200 MeV bombarding energy [Meyer et al., Phys. Rev. C Z7_, 459 (1983)]. We have therefore decided to devote the beam time remaining for Expt. 221 to a study of cross- section and analysing power angular distribu­tions of 12C(p,p) 12C at 250 MeV, filling a gap in the available experimental informa­tion. This measurement is currently inprogress.Experiment 223The (p,2p) reaction and the momentum distributionof the deuteron (C.F. Perdrisat, William & Mary;W.T.H. van Oers, Manitoba)0 , m. (deg)Fig. 34. The results of the present experi­ment: the cross section of 12C(p,p)12C atTp = 300 MeV. The solid curve is an optical model fit to the data using a double Woods- Saxon shape for the real central potential. The dashed line includes in addition the ef­fect of a coupled channel due to intermediate A isobars. The coupling strength has been adjusted to account for the back-angle en­hancement of the cross section observed at 200 MeV.The principal interest in studying the (p,2p) and (e,e'p) reactions for light nuclei is the determination of the single-nucleon momentum distribution (SNMD). In the case of the deu­teron excellent electron data now exist for a range of internal momenta up to 650 MeV/c. These electron data determine the SNMD after several corrections have been applied; most important among them is the meson exchange current (MEC) correction. For the 2H(p,2p)n reaction there is an additional strong-inter- acting particle in the final state, an added difficulty, but no MEC effect. With the strong-interacting probe we expect other phe­nomena to be important, the most interesting being isobar excitation (IE). It has long been suspected that a conspicuous difference between the electron and the proton results,31namely the much larger values found for the SNMD in proton experiments at large internal momenta, had its origin in IE. A recentcalculation of IE contributions by Yano shows that for large neutron recoil momenta in the ^H(p,2p)n reaction the cross section corre­sponding to a pion exchange graph which in­cludes excitation of the A(1232) is much larger than the plane wave impulse approxima­tion (PWIA) pp and pn contributions.In 1983 new data pertaining to the isobarregion were obtained in the first phase of Expt. 223. The results are about to be sub­mitted for publication, simultaneously with the calculations of Yano. In the A excitation region the invariant mass of the two protonsMpp ~ m^ +mfj; this region can be reached byincreasing the proton angles from their quasi-free (QF) values, which simultaneously increases the neutron recoil momentum. The data include the equal angle pairs of 50°, 52°, 57° and 66° with equal energies of the two protons. The neutron recoil momenta in the A region have values between 200 and 650 MeV/c.The choice of symmetric kinematics has tech­nical advantages and leads to final states which have the neutron momentum parallel to the incident proton beam direction. We have also obtained in 1983 data for symmetric angles of 41.5°, which is a QF point corre­sponding to zero recoil for equal proton en­ergies. Unequal proton energies lead then to neutron momenta perpendicular to the incident direction; the data points have better than 1% statistics and will give the SNMD with high accuracy.The 1984 second phase of Expt. 223 had sever­al objectives:a) obtaining additional data in the QF region with smaller solid angles defined by counters rather than by software cuts (high accuracy data)b) obtaining new data near the QF region with good statistics (symmetric angles of 38°, 44° and 47°) for a detailed study of the SNMD up to 150 MeV/c neutron momentumc) obtaining new data near the QF region for asymmetric kinematics (the QF point at 30°-53.75°), 30° with 30°, 37°, 44°, 53.75°, 61°, 68° and 75°d) obtaining new data away from the QF region for asymmetric kinematics, 14° with 33°, 41.6°, 53.75°, 62°, 73° and 85°The data are presently being analysed atCalifornia State University at Los Angeles and College of William and Mary. The infor­mation we are seeking can be phrased as fol­lows :1) Does the SNMD extracted from the data with the help of the PWIA agree both in shape and magnitude with current deuteron wave func­tions in the small neutron recoil momentum region [data from a), b) and c)]? Previous experiments have given SNMDs too small by 10 to 20%. For this purpose very careful cali­bration and accounting of the efficiency of the data acquisition system are needed. This analysis is approaching completion.2) Is A excitation the only important process in addition to the IE pp and pn processes? (We know that rescattering is of relatively minor importance in this energy range.) Toanswer this question we have obtained data in the large recoil momentum region, keeping the invariant mass in all three pairs of nucleons away from the (m^ +mjj) region. One point which is particularly interesting is the equal-energy sharing situation, where the relative energy is the same for all three pairs of nucleons. We expect the SNMD obtained under these conditions to be closer to that ob­tained in electron scattering. By itself such a result would be further evidence that the enhancement seen in the first phase of theexperiment is indeed caused by the excitation of the A. The data from c) and d) will pro­vide information concerning this question. The analysis of these data is in progress.Experiment 224Inclusive scattering o f pions from very lightnuclei at 100 M eV  (I. Halpern, Washington)Last February we completed a measurement of inclusive inelastic scattering of 100 MeV pions from ^H, H^, 4He and ^N. One motiva­tion for studying the inclusive scattering of pions from these targets was to learn about the interplay of quasi-elastic scattering and pion absorption in the very simplest nuclei.The experiment was performed on the Mil chan­nel using the QQD spectrometer and a high pressure (100 atm) gas target. The use of such a target makes it possible to determine relative cross sections for the different targets with good precision. ir+ scattering data were taken at five laboratory angles (40°-125°) and some ir- data, also at 100 MeV, were taken on ^He and ^He at 60°, 100° and125°. The tt“ comparison should help deter­32mine the role of absorption in setting the magnitude of the scattering cross sections. Data were collected in four momentum bites of the spectrometer in order to cover up to 85 MeV in excitation energy.We have finished a preliminary analysis of both the tt+  and i r ~  data. The shapes of the scattered pion energy spectra for all four targets are roughly as expected from a quasi­elastic scattering picture. The absolute cross sections (obtained from normalization to our hydrogen data and known ir+-proton cross sections) of the ir+-ltHe inelastic, tt+-2H elastic and ir_-3He elastic scattering compare well with the results of Baumgartner et al., [Nucl. Phys. A399, 451 (1983)],Gabathuler et al. [Nucl. Phys. A350, 253(1980)] and Kallne et al. [Phys. Lett. 103B, 13 (1981)], respectively. The ir+ inelastic scattering angular distributions for 2H, 3He and 4He in the back hemisphere have very nearly the same shape as the free r+-proton angular distribution at this energy. In Fig. 35 we show the inelastic 4He angular distribution. The ratio n of the inelastic angular distributions to the free ir+-proton angular distribution (fit to 75°<0iab<125°) 2H, 3He and ^He is 0.73±0.04, 1.30±0.07 and 0.85±0.04, respectively. The 14N inelastic angular distribution is somewhat steeper with angle than the free ir+-proton distribution, in accord with our previous findings in the mass region A=12-208 [Nuclear Physics Labora­tory Annual Report, University of Washington(1982) pp. 98-114); Nuclear Physics Labora­tory Annual Report, University of Washington(1983) pp. 52-56].Our goal is to understand our findings on these simple systems quantitatively in terms of shadowing effects, nucleon momentum dis­tributions and the pion absorption probabili­ties .Experiment 234Studies o f the A(p,n )A + 1  reaction (R. Bent, IUCF; G.J. Lolos, Regina)Although the upgraded MRS has progressed to the point that ( p , t )  experiments can be car­ried out with good resolution and reasonable beam intensities, it has proven very diffi­cult to use the facility effectively when resolution of ~120 keV FWHM is required, due to the poor quality of the extracted beam when high beam current operations are in effect for meson hall users. (The (p,r ) ex­periment requires beam quality of ~1% Ap/p soLAB ANGLE (degrees )Fig. 35. Angular distribution of 100 MeV ir+ inelastic scattering on ‘'He. The line is the free ir+-proton angular distribution multi­plied by the factor 0.85.that an effective 20 cm/% dispersion can be attempted. So far attempts to obtain such conditions have failed.The successful conditions under which Expt. 218 (with similar but less stringent requirements) ran in November are encouraging and we hope that the January 1985 (p,r-) run against 30 pA of machine operation will pro­vide the necessary quality of beam to proceed with Expt. 234.Experiment 243Energy and angle dependence o f the 6Li(n+, 3He)3He reaction(G.J. Lolos, Regina)In the summer of 1984 Expt. 243 received 26 shifts at Mil that proved sufficient to com­plete the experiment. The detector configur­ation remained unchanged from the 1983 TRIUMF Annual Report and consisted of six tele­scopes, of four detectors each, arranged in three pairs of arms to detect the 'back-to- back' He nuclei.Data were obtained at 03jje = 15°,30°,45°, 60°,75° and 90° for incident pion energies of 60, 80 and 100 MeV. A more restricted angular distribution (at 15°, 45° and 75°) was mea­sured for Ttj+ = 140 MeV. The 3He identifica­tion was clean, as can be seen from Fig. 36. The angular distributions of the differential cross sections have been extracted for the 60 and 80 MeV data and are shown in Figs. 37(a) and 37(b). In the same figures the theoreti­cal calculations by Germond and Wilkin, [j.33Fig. 36. Two-dimensional 3He spectrum takenat 30° for the two arms.1,,+ = 80 MeV for the forward of1000 r100.. a) T7T =6 0  MeV: - 1  ,4 9  .3 MeV ____ _ __---------------*-. + *<J> 59.3 MeV ^ 4 9 .3 M e V: -*6 0 M e V •tPhys. G10, 745 (1984); private communication (1984)] are shown for two types of nuclear potentials. The 100 MeV results in Fig. 37(c) are pr liminary as are the total cross sec­tions in Fig. 38, but they serve the purpose of presenting the overall energy dependence of the total cross sections and the angular distributions of the differential cross sections.Where direct comparisons with older published results are possible the older TRIUMF results [Lolos et al., Phys. Lett. 126B, 20 (1983)] are a factor ~1.7 lower than the present results at 60 and 75 MeV, while the 90 MeV result is in good agreement with the present experiment as are the results from Saclay [Le Bornec et al., Phys. Lett. 114B, 311 (1982)]. The LAMPF result [Barnes et al., Nucl. Phys. A402, 397 (1983)] at 59.3 MeV is a factor ~2 higher than the results in the present work. The energy dependence of the cross section is a simple exponential drop with increasing-OcOUJCO2oI 0 2  --------------------------- i---------------------------- ‘---------------------------- 1--------------------- —5.7 5 0  5 .7 8 0  5.810 5 .8 4 0  5.871Ec.m<GeV>Fig. 38. Total cross-section energy depen­dence. This work ■ ; Orsay □ [Le Bornec et al. [op. cit].Fig. 37. Angular distribution of da/df2 for a) T„=60 MeV, b) ^=80 MeV, c) ^=100 MeV. This work ■; Saclay A [Le Bornec et al., op. cit.]; TRIUMF (early) x [Lolos et al., op. cit.]; LAMPF 0 [Barnes et al., op. cit.]. For c), data points are as defined in the, and a linear fit of the data, includ­ing the lower energy ORSAY data [Le Bornec et al., Phys. Rev. Lett. bl_, 1870 (1981)], gives good results. Although the Germond and Wilkin model reproduces the small angle da/dfi, the calculated angular distribution does not re­produce the experimental features.34Experiment 246Pion double charge exchange on 180(R.R. Johnson, TRIUMF/UBC)The QQD spectrometer was used on M13 to per­form 180 (ir+,ir-) 18Ne at 50 MeV. The double charge exchange reaction has been proposed as a candidate for observing quark nucleus ef­fects. Miller [Phys. Rev. Lett. 53, 2008(1984)] has compared scattering calculations where long-range effects were described by more conventional pion exchange mechanisms but interactions of <~1 fm were presented in a quark framework. His calculations closely follow the double charge exchange reaction on performed by the TRIUMF TPC [Navon et al., Phys. Rev. Lett. 52, 105 (1984)]. A double charge exchange experiment at LAMPF [Leitch et al., preprint LAUR 84-2754 (1984)] extended the TPC measurements to smaller angles and when pion absorption effects are included the quark-at-short-distance, pion- exchange-at-long-distance type of calculation does reproduce the data. However, a more con­ventional calculation using the A-hole model [Karapiperis and Kobayashi, preprint SIN PR- 84-14 (1984)] does reproduce the ^ C  data as well. The QQD 180 double charge exchange dataJD=L<_>UJtotoCOoi rozz 0 .4Q LbJ0 .1G .M ILLER CALCULATIONS WITH ABSORPTION  WITHOUT ABSORPTIONPRELIMINARY  CROSS SECTIONS1i80 ( tt+ ,T7 -)  l8Ne DIAS AT 5 0  MeV_L_ _L_2 0  4 0  6 0  8 0  1 0 0  120e,c.m.Fig. 40. Preliminary cross-section data for the double charge exchange reaction on 180 at 50 MeV. The curves are calculations by Miller that include 6-quark effects.COUl>Ll ILl .oo|DIAS 18, .18.'0 (7r +, 7r  )l,JNe TV+ = 5 0  MeV# L A B  = 5 0 M e VON LINE SPECTRUM NO CUTS____________(a)A p / p > (b)Ll It— Itr _i I -  u-COS P E C T R O M E T E R  F L IG H T  T IM E  OF RAW  E V E N T S  FO R D O U B L E  C H A R G E  E X C H A N G E_________________________  E L E C T R O N  BAND —> *  V -  *  A »  "  •PIO N  B A N D -VL^ DIAS& P ' P 0Fig. 39. (a) On line spectrum of doublecharge exchange from ^80; (b) two-dimensional spectrum of momentum and spectrometer time of flight for spectrometer 50 MeV must also be explained by the same type of calculation that has been applied to 14C but now the nuclear structure contribu­tion has been changed by changing the target nucleus.The spectrometer was used with the 50 MeV M13 pion beam in the same configuration as elas­tic and inelastic scattering experiments have used in the past [Sobie et al., Phys. Rev. C 30, 1612 (1984)] but with the spectrometerpolarity reversed and magnetic fields ad­justed to place the double analogue transi­tion in the centre of the exit wire chambers. Figure 39(a) shows a spectrum with no cuts made to the data. Since the spectrometer configuration allows a projection to the scattering target to be made, a high signal- to-noise spectrum is immediate. The maintarget-located background is associated with electrons. Figure 39(b) presents a two- dimensional histogram of time-of-flight and momentum characteristics of the events. The electrons are clearly removed by a time-of- flight cut and a Cerenkov veto was not em­ployed for these runs.Figure 40 summarizes the preliminary double charge exchange data at 50 MeV. For eachangle that was measured, an *80 elastic35scattering cross section was measured and we are now using that data and earlier data [Tacik et al., Phys. Rev. Lett. 52^ 1276(1984); Barnett et al., preprint (1984)] to confirm the acceptance of the spectrometer for this experiment. These preliminary data do agree with calculations by Miller for the 180 double charge exchange reaction [private communication (1984)].Experiment 272Transverse spin-flip probabilities in 24M g (p j? )and 48(p,p') (O. Hausser, Simon Fraser)In (p,n) and (pp') reactions at low momentum transfer only about half the AS = 1 spin-flip strength expected from the shell model is concentrated in strong peaks associated with 0 ho> excitations. Bertsch and Hamamoto have suggested that the tensor component of the nucleon-nucleon force may distribute the missing strength over a ~30 MeV wide energy interval. Experiment 272 is designed to locate distributed AS = 1 strength in (pp') reactions at extreme forward angles through measurements of Snn, the transverse spin-flip probability. The key element of the experi­ment is a new focal-plane polarimeter for the MRS which is described elsewhere. The polar­imeter will be operational in early 1985.4 0 0UJ<IoN(/>ZDoo320160l2 C(p,p') Q =3°2 0T 4 .4 4  7.65♦ J_4___a)Ep=300 MeV0 400  2000  3600MOMENTUM (CHANNEL NUMBER)2400 1000 2 00 0  3000MOMENTUM (CHANNEL NUMBER)Fig. 41. Missing mass spectrum (a) for *2C observed at 9=3° and Ep=300 MeV, (b) for 24Mg observed at 9=3° and Ep=250 MeV.First test experiments with the upgraded MRS in the small angle configuration (SAC) have been carried out in September. Particles scattered in the beam blocker at the entrance to the MRS cause an intense background which were excluded by ray tracing back to the tar­get and by imposing numerous software condi­tions. Very clean spectra resulted at scat­tering angles as small as 3° [see Fig. 41(a) and (b) showing a *2C spectrum at Ep = 300 MeV and a 2 Mg spectrum at Ep = 250 MeV] The resolution was 140 keV for *2C (target thickness = 48 mg/cm2) and 160 keV for 24Mg (target thickness = 48 mg/cm2). A thin veto scintillator was introduced near the focal plane to scale down the extremely intense elastic peak. The 1.37 MeV first excited state in 24Mg could be observed without losses whereas the elastic peak was drasti­cally reduced.Approximately four shifts of polarized beam time were used to measure cross sections and analysing powers in 2ltMg(pp') at angles be­tween 3° and 15° in 2° steps. As expected the 1+,T=1 state at 10.71 MeV excitation was the dominant peak at the most forward angles. At 9c.m .=3.16° a cross-section value (da/dfi)c.m< =1.65±0.06 mb/sr was obtained. Using the Ham­burg effective NN potential [von Geramb, in The Interaction between Medium Energy Nucleons in Nuclei-1982, AIPCP #97 (AIP, NY, 1983), p. 44] and wave functions by Chung and Wilden- thal, we calculate (da/df2)tiieor=2.4 mb/sr and a quenching factor of 0.7. The reliability of the extracted cross sections was tested with H(p,p) scattering and found to be better than ±7%. Further analysis of these data is in progress. The results for the continuum above Ex=12 MeV are used to assess the run­ning time required for the Snn measurements.36RESEARCH IN CHEMISTRY AND SOLID-STATE PHYSICSExperiment 147Formation and reactivity o f positive muonsand muonium in gases (D. Fleming, UBC)Muonium is formed during the slowing-downprocess of the y+ as a result of cycliccharge exchange. The fraction of observable muonium at thermal energies (PM) is deter­mined by the relative magnitudes of the charge exchange cross sections forming muoni- um (a1Q) or returning the free y+ (a01), in competitipn with hot atom or ion reactionsplacing the muon in a diamagnetic environment (Pp). If the time between collisions of the Mu atom is >l/v0, where v0 = 4463 MHz, the U+-e- hyperfine interaction frequency, then significant depolarization of the p+ occurs, resulting in reduced amplitudes for Pp and P^ with a corresponding nonzero Pp; conversely Pp+O in the limit of 'high' (~1 atm) pres­sures. At thermal energies the relaxation of the pSR signal due to the interaction of the muon ensemble with its environment gives rise to the study of reactive collision phenomena, of either y+ molecular ions (Pp) or of the Mu atom itself (P^) • Such studies are of considerable interest to the chemical physics community at large.Muonium formation and the missing fraction in vapoursWe have studied the pressure dependence of Mu formation in a variety of molecular vapours [Arseneau, M.Sc. thesis, UBC (1984); Arseneau et al., J. Phys. Chem. 8iB, 3688 (1984)] in order to (i) compare with similar data avail­able from proton charge exchange studies and (ii) compare the observed polarizations Pp and Pm with corresponding work in condensed media [Walker, J. Phys. Chem. U5, 3960(1981)]. Since the ionization potential of all of these vapours is less than that of the Mu atom (13.6 eV), at low energies and hence 100% Mu formation should be ex­pected at high pressures. In every case though PM ~ 80%, except in CCJl^  where PM ~ 50%; the differences, reflected in corre­sponding increases in Pp, are attributed to hot atom (or ion) reactions likely occurring after the cyclic charge exchange regime (from -20 eV to kBT).Figure 42 shows the data obtained (Pp+PM) in CC l^ as a function of pressure; the fit is from a simple model for lost polarization [Arseneau, MSc. thesis; Fleming et al., Phys.Rev. A 26, 2527 (1982), Hyp. Int. 17-19, 655(1984)]. Much more sophisticated treatments [Turner and Senba, Phys. Rev. A 29, 2541(1984)] yield qualitatively similar results. Unlike all the other vapours we have studied C C l k (and possibly CHC£3) exhibits a true missing fraction in that Pp is not asympto­tic to zero at high pressures (Fig. 42, Table VI). The origin of this lost fraction is not understood but suggests the formation of unobserved muonic radicals.Table VI compares representative cases of the high pressure asymptotes for P^ and Pp with the corresponding polarizations seen in the liquid phase for a variety of molecular vapours. It is noteworthy that in every case PM (vapour) > PM (liquid), with Pp exhibiting the opposite trend; in addition Pp « 20% in the condensed media. This marked difference in the distribution of muon polarization between the vapour and condensed phases strongly suggests that different mechanisms for Mu formation are operative in media of widely differing densities; in particular, it is likely that a radiation chemistry 'spur' mechanism dominates Mu formation in condensed media [Percival et al.; Miyake et al., Hyp. Int. 17-19, 721, 727 (1984)]. On the other hand, the basic (two-body) charge exchange/ hot atom (ion) model which nicely explains the vapour phase results [Arseneau et al., J. Phys. Chem. 88, 3688 (1984) and Fleming et al., Can. J. Chem., submitted] at low pres­sures may become completely altered at higher and higher pressures (densities) where many-Pressure ( atm at 425 K )Fig. 42. PM + Pp vs. pressure for carbon tetrachloride. This indicates a true (high pressure) missing fraction.37Table VI. Comparison of muon fractions in molecular vapours3 with those in condensed mediaMolecule Medium Pd PM Plh 2o g 10 + 2 90 + 2 0A 62 + 2 20 + 2 18 + 3CHjOH g 15 + 2 85 + 2 0A 62 + 2 23 + 2 15 + 3C6H14 g 24+ 2 76 + 2 0A 63 + 3 18 + 3 19 ± 4C6H 12 g 25± 2 75 + 2 0U  -IA 66 + 3 16 + 3 18 + 4Si(CH3)lt g 20 + 2 80 ± 2 0A 58 + 4 20 + 3 22 + 5cc\ g 32 + 3 38 + 2 30 + 4A 100 0 0aGas phase results correspond to high pressure: ('~1 atm)asymptotesTable VII. Comparison of Mu* iand T* hot atom yieldsi andreactivity integrals in some alkane vapours (RH)RH T*(Yield) Mu*(Yield) I'p* iMu* It*/Imu*CH^ 0.50 0.12 0.70 0.09 7.9C2Hg 0.65 0.19 1.5 0.25 6.0c3h8 0.66 0.21 2.1 0.42 5.0n - C ^ Q  n-CcHio0.68 0.21 2.3 0.45 5.10.69 0.15 -2.4 0.33 -7.3n-C6H11+ 0.72 0.24 -2.5 0.52 -4.5Table VIII. Muon molecular ion reaction [NeMu+]rates kc for [HeMu+ ]*,Reaction kexp(10 10cc s L) ktheory(10_1°cc s_1)[HeMu+ ]* + Xe 12 ± 4 23+ CH^ <0.02 21+ nh3 60 ± 20 36+ CHoF 29 ± 5 37+ n 2o 20 ± 2[NeMu+ ]* + Xe 4 ± 1 11+ CH^ <0.02 12+ nh3 26 ± 3 21+ CHoF 9 ± 3 20+ n 2o 11 ± 238body effects may play an Important role. Additional experiments through the critical region of selected vapours are planned.Muonium hot atom chemistryAs part of its thermalization process, muoni­um with kinetic energies »  kgT (i.e., 'hot') can undergo certain chemical reactions with much larger cross sections than those pre­vailing at thermal energies. These processes are manifest in the magnitude of the diamag­netic fraction Pq or hot atom 'yield' which can be interpreted in terms of a 'reactivity integral' I defined byr  e2I = / a(E)f(E)dEwhere o(E) and f(E) are the reactive cross section and distribution function of hot atoms in the energy interval E2(~20 eV) - Ej(>kgT), respectively. We have recently compared hot Mu(Mu*) with hot tritium (T*) yields for a variety of alkane (RH) modera­tors in order to establish the importance of dynamic mass effects in hot atom chemistry [Fleming et al., Can. J. Chem., submitted].Table VII presents the hot atom yields for both Mu* and T* with a variety of RH mole­cules as well as the corresponding reactivity integrals I. This is really the first de­tailed comparison of the effects of isotopic mass on hot atom reactivity at the most sen­sitive end of the mass scale - a factor of 27 between Mu and T! In general, the ratio of hot atom yields for T*/Mu* are “3/1, which translates into a ratio of reactivity inte­grals It*/1mu* w 6/1, calculated according to to the phenomenological prescription of Wolfgang-Estrup theory. There are no theore­tical calculations of hot atom reactivity for the molecules of Table VII. Current calcu­lations comparing Mu* and T* reactivity with F2 [Connor et al., Mol. Phys. ^6, 1231 (1982) and work in progress] give only ~2 for the ratio It*/1mu*> not consistent with the fac­tor 6 seen in the present study. It is noted that these ratios are likely to be less sen­sitive to details of the (unknown) potential energy surfaces than the actual cross sec­tions themselves.Muon molecular ions and ion-molecule reac­tionsAs outlined in last year's Annual Report [see also Fleming et al., Chem. Phys. 82^ , 75(1983)], the diamagnetic signal from positive muons stopped in the noble gas moderators He, Ne, Ar is due to the formation of the p+ mo­lecular ions [HeMu+ ]*, [NeMu+ ]*, [ArMu+]*, in ro-vibrational excited states, in analogy with their well known protonic analogs. The pSR signal due to these species is very long lived, T2 ~ 50 ps. Upon the addition of some reagent X (with an ionization potential <13.6 eV), thermal charge exchange causes depolarization and hence a relaxation of this signal, in competition with muon transfer, according to, e.g. with NeMu+ ,' k - j r  Mu + X+ + Ne[NeMu+ ]*+XXMu+ + Newhere kc and kt are the rate constants for charge exchange and muon transfer, respec­tively. This competition gives rise to two relaxations, a fast component due to kc and a slow one which is not directly related to kt (in fact, kc/kt “ 1). We have now studied kc at room temperature for a variety of reagents X(Xe, Kr, NH3, CH4, C2H6, CHjF and N20) in both He and Ne moderators as well as Xe, NHj and CH^ in Ar moderator. There is no relaxa­tion seen in Ar regardless of reactant, dem­onstrating that the [ArMu+ ]* molecular ion has too great a binding energy for charge exchange to be exothermic. There is also no relaxation with CH^ and likely also not with C2H6 in either He or Ne moderators but there is a fast relaxation with all the other reac­tants in both He and Ne, including Kr (at least in Ne). The latter result is actually a surprise since Mu formation with Kr should be endothermic, even for a bare muon.Figure 43 shows recent data for the relaxa­tion rate A vs the concentration of X, [X], for the [NeMu+]* + Xe reaction, in comparison with earlier published work [Fleming et al., Chem. Phys. 82^ , 75(1983)]; the agreement with the earlier data is excellent. The more recent data, however, goes to much lower con­centrations, demonstrating a pronounced curv­ature in A at these concentrations which was not established before. All reactants X extrapolate to the same kind of dependence on concentration at low [X] but it is not yet clear whether the high concentration depen­dences are the same for all X. Some seem to indicate a plateau effect, reminiscent of a unimolecular type of reaction mechanism, whereas others exhibit a linear dependence at high concentrations; indeed, within errors, all the high concentration points can be fit to a linear dependence (see also Fig. 43)39Xenon C oncentration (lO ^y c m 3 )Fig. 43. The pSR relaxation rate of [NeMu+]* vs. [Xe]. The curvature at low concentration is not understood. The straight line fit gives an estimate of the charge exchange rate constant, kc.from which the rate constant kc can be determined. These values for k~ are shown in Table VIII for both the HeMu^ " and NeMu+ ion-molecule reactions, where they are com­pared with the theoretical ones calculated from the simple Langevin theory for those molecules with zero dipole movement (Xe, CH^) or from the average dipole orientation (ADO) theory for those molecules with a permanent dipole moment (CH3F, NH3). The errors on the experimental numbers are large (~±25%), re­flecting present uncertainties about the reaction mechanism. Given this level of uncertainty the agreement between theory and experiment is good, certainly 'state-of-the- art' as seen in a wide variety of ion mole­cule reactions, with the notable exception of CH4. The lack of any observable reaction in this case is certainly puzzling and suggests that the competitive process of (unobserv­able) muon transfer (kj-) dominates over charge exchange. It is noted that in every case the experimental rates for the HeMu+ ion are faster than those for the NeMu+ ion, as expected. Further work will include a mea­surement of the T dependence of these muon ion-molecule reaction rates.Experiment 150Muonium in ice (P. Percival, Simon Fraser)Our interest in ice developed out of the pro­tracted study of muonium in water and aqueoussolutions. Landmarks in this area include the first detection of muonium in water [Percival et al., Chem. Phys. Lett. 39^ , 333 (1976)], the proposal and investigation of radiolysis effects [Percival et al., Chem. Phys. 32, 353 (1978)], and demonstration of the role of hydrated electrons in loss of muonium spin polarization [Percival et al., Chem. Phys. Lett. 91, 1 (1982)]. These advances directly contribute to the goal of understanding the factors which determine the distribution of muons amongst the primary chemical species formed at the end of the track in condensed matter. An additional application of muonium in ice is the study of hydrogen atom diffu­sion. Experimental factors have precluded the detection of free H atoms in pure ice in the temperature range 50-160 K, and there has been no work reported on H in ice monocrys­tals. In contrast, we have been able to detect Mu in ice from 10 K up to the melting point. Furthermore, our study of single crystals led to the discovery of a small axial anisotropy in the hyperfine interaction [Percival, Chem. Phys. Lett. 93_, 366 (1982)].Progress this year has been as follows:1) A combined theoretical and experimental analysis of muonium spectra in polycrystal­line ice resulted in full understanding of the apparent discrepancies between the old (SIN) data from polycrystals and the recent work performed with single crystals. Careful measurements of muonium and diamagnetic sig­nal amplitudes revealed that, contrary to earlier belief, there is a missing fraction of muon polarization in ice at temperatures above 200 K. It is thought to arise from nonreactive spin exchange encounter between muonium and mobile hydrogen atoms. This work was written up as a major full-length paper [Percival et al., Chem. Phys. (in press), TRI-PP-84-27].2) The acquisition of a helium refrigerator allowed us to extend our measurements of muonium and diamagnetic signals in ice from 100 K. down to 10 K. To date we have ascer­tained that the diamagnetic fractions in H20 and D20 do not keep falling (as they do from 150-100 K) but level off around 0.3 and 0.2, respectively.3) The field dependence of muonium frequency splitting has been measured for Mu in D20. Analysis of the results gives a value of 4590 ± 10 MHz for the isotropic hyperfine frequency. Since this is a little higher than the free vacuum value (4463 MHz) it implies401,511.0 - L°g am0.5- 0.0 - -0.5 - - 1.0 --1.5 H------- 1------ 1-------- 1----- 1-------- 10 0.5 1 1.5 2 2.5Log (1000/T)Fig. 44. Muonium spin relaxation rates for single crystals of H20 (open symbols) and D20 (closed symbols) in various transverse mag­netic fields.that muonium is in a slightly repulsive po­tential in the ice lattice.4) Measurements of muonium spin relaxation rates have continued, and the full tempera­ture range has now been covered for both H20 and D20 (see Fig. 44).T im e  in tts e c  (3 .1 25  n s e c /b in )Fig. 45. Diamagnetic muon asymmetry in cry­stalline CC£4 at 213°K fitted to a two-relax­ation (Gaussian + exponential) function.a single comprehensive process. In the cry­stalline phase, however, the diamagnetic muon asymmetry has two components, distinguishable by different relaxations as shown in Fig. 45. The faster relaxation fits better to a Gaus­sian function than to an exponential. Its amplitude and relaxation rate vary with temp­erature. In concert with the finding of muonium in gaseous (Expt. 147), thisshows that the muon chemistry of CCJt^  is not as simple as previously supposed.Experiment 157Muonium chemistry in condensed media(D.C. Walker, UBC)Muon spin rotation studies on M20 have con­tinued to examine muonium atom reactivity to­wards aromatic solutes in dilute aqueous sol­ution at room temperature. The results have been interpreted through the Hammett rela­tionship. Addition reactions giving cyclo- hexadienyl-free radicals produced a small positive p factor using established a sub­stituent coefficients for a series of benzoic acids. A positive p signifies nucleophilic character. It contrasts the slight electro- philic nature shown by muonium's heavy iso­tope H. It remains to be seen whether it is Mu-radical selectivity here (Expt. 262) that distinguishes Mu from H, or whether it is a subtle kinetic isotope effect (mass/quantal difference), or whether it is tied to the protonic character of the medium.The other sequence of experiments pursued further in 1984 concerns CCA^, the compound that has invariably been used to calibrate the polarization. It was presumed to give a 100% yield of muons in diamagnetic states byExperiment 161 Amorphous spin glasses (J. Brewer, UBC)The amorphous alloys a-PdgQP20 (nonmagnetic) and a-Pd/sFe 5P20 ( spin glass) were investi­gated for comparison with the similar amor­phous spin glass a-Pd/5Fe5Si2Q studied earli­er. The substitution of P for Si in these alloys allowed us to check the monmagnetic version for p+ diffusion as manifested in motional narrowing of p+ relaxation by P nuclear dipoles. We found no evidence for fast diffusion; the spin glass results can therefore be interpreted straightforwardly.Below the spin freezing temperature Tg the local field B can be characterized by its Edwards-Anderson order parameter Q, a measure of the fraction of the randomly oriented B which is static as opposed to fluctuating with a correlation time t .For a-Pd/5Fe5Si20, Tg = 15.5 K, whereas for a-Pd/5Fe5P20 Tg = 10 K. However, when plot­ted in terms of the reduced temperature t = T/Tg, both Q(t) and x(t) are virtually identical for the two samples; furthermore,41x(t) has the same shape as for all the spin glasses so far investigated with ySR, namely x(t) ~ (1-1/t)2, indicating that this scaling universality extends even to amorphous systems.A sample of Pd8gFe2Mng from Japan was also studied using ZF-pSR from 40 K (where it is paramagnetic) down to 15 K (where it becomes ferromagnetic) and through 6 K (where it ex­hibits the remarkable property of becoming a re-entrant spin glass).Experiment 191Muonium relaxation on silica surfaces(J. Brewer, UBC)Our experimental study of muonium adsorption, trapping and diffusion on the surfaces of 7 nm Si02 powder grains was completed in 1984 and will be described in D.R. Harshman's forthcoming Ph.D. thesis. The final experi­ments on zero field (ZF) and longitudinal field (LF) muonium spin relaxation (MSR) were intended to identify the local field distri­bution and relaxation mechanism(s) at the muonium surface sites. Interpretation of these results has proved to be a subtle but interesting problem: the ZF spin relaxationfunction G(t) was found to be similar to that expected for Mu in random local magnetic fields (RLMF) from randomly interspersed proton dipole moments in surface OH groups, but the 1/3 tail recovery at long times was missing even at low T, suggesting persistent quantum diffusion of the Mu atoms on the surface. Such motional effects failed to manifest themselves as expected, however, in the LF relaxation; a Mu atom hopping among sites will always be relaxed, even in a de­coupling LF, yet we found a fraction of the Mu polarization to be well stabilized by LF of a few gauss. We conclude that while the hydroxyls clearly play an important role in trapping and relaxing muonium on silica sur­faces, a major part of the relaxation mecha­nism is due to random hyperfine anisotropies (RHFA) caused by the distortion of the Mu electronic wave function as it sticks to the surface. This relaxation mechanism, and its response to dynamics and/or LF, is now being modelled in detail.Last year we reported that Mu trapping at 6 K on OH-depleted surfaces was dramatically enhanced by adsorption of half a monolayer of helium. This effect is independent of pack­ing density and of which He isotope is used. We now suspect that the filling up of the empty OH sites by He atoms forms bridges al­lowing freer motion of Mu, which then passes through a sort of percolation threshold at half coverage, allowing the Mu to find the deep trap sites more easily; at higher He coverage the deep traps themselves are filled in and rendered innocuous.A 10% monolayer of argon was deposited on the silica surface (baked at 100 C) in an attempt to alter the trap concentration. A marked change in the TF Mu relaxation rate X was observed, indicating a change in the surface characteristics analogous to that described above.Experiment 219The chemistry o f pionic hydrogen atoms(D.F. Measday, UBC; D. Horvath, Budapest)The role of hydrogen in the molecular, atomic and eventual nuclear capture of a it -  stopped in a hydrogenous material is complicated and, as yet, not fully understood. The aim of Expt. 219 is to study the processes involved by comparing the ir-p capture probabilities in various compounds, or in the same compound under different conditions, by measuring the y-rays from the reaction:p+ir- •> n + ir° (60% branching ratio)L >YYusing the two large Nal detectors, TINA and MINA.The atomic capture of a ir~ by hydrogen is in­fluenced by the hydrogen's chemical environ­ment. The TT~p system is small, neutral and weakly bound which allows it to pass close to other nuclei, where due to the deeper poten­tial well of a higher Z element it is predis­posed to transfer the it -  to the nuclei, re­ducing the ir“p nuclear capture rate.In the July experimental run for Expt. 219 the relative ir“p capture probabilities of normal and deuterated methanol were measured over a range of temperatures to see if there is a relationship between the degree of hyd­rogen bonding of the -OH group and the ir-p formation rate for that hydrogen atom. Such relationships have been seen at Dubna in the cases of H20 and NH8 [HorvAth et al., Chem. Phys. Lett. 87_, 504 (1982)]. In addition a series of measurements were made on materials of biological interest.Interest in the pion chemistry of organic materials has been revived by the use of pion radiotherapy, where the effective radiationdose to the tumour may depend on the relative tr“ capture probabilities of the different elements present. The University of Surrey, England has obtained funding to investigate the implications of pion chemistry for pion dosimetry. In November 1983 D.F. Jackson sent a postdoctoral fellow, J.R.H. Smith, to TRIUMF for a year to work with D.F. Measday's group on the role of ir-p capture in organic and related materials. A parallel study of pionic X-ray measurements was carried out by a Surrey Ph.D. student on the biomedical line (M8).Beam line Mil was used to deliver a beam of 150 MeV/c negative pions. A stop in the tar­get was defined by the B ^ B ^ ^ ^  coincidence. Counters B3 and were both made of deuter- ated plastic scintillators to reduce ir-p background. TINA and MINA were used in coin­cidence to detect the ~70 MeV y-rays from the it0 decay. Signals from scintillation coun­ters placed in front of Nal detectors were used to identify, and eventual reject, the charged particle events. A graphite degrader of variable thickness directly upstream of B3 was used to optimize the incident pion energyFor D. Horvath's part of Expt. 219, i.e.,studies of hydrogen bonding effects, measure­ments were carried out for CH30H, CD30H and CH30D at -120°, 20°, 100°, 200° and 250°C. H20 was measured at 20° and D20 at -150° and 20°C to calibrate the experiment. The meth­anol was sealed inside a sample cylinder made of monel with a wall thickness of ~1 cm. The cylinder was then placed inside an insulated aluminum box which could either be heated by hot air from a furnace or cooled by liquid nitrogen. B^ was not used for this part of the experiment as pions stop entirely in the methanol inside the target.Hydrogen bonding is strongest at low tempera­tures in the fluid states and is constant in the solid state, causing the electron density near the hydrogen, and hence the ir-p forma­tion probability, to be at a minimum. As temperature increases, bonding decreases, in­creasing the electron density and hence the ir-p formation probabilities, perhaps by as much as a factor of 2. There should be no, or very little, change for the CH30D ir“p for­mation probability with temperature.The CD30H measurements did show some increase in the ir-p signal with temperature. Using TINA and MINA in coincidence produced a prac­tically noise-free signal. Neutron events were discriminated against by time-of-flightinformation. The ir° signal could be easily distinguished from the background by imposing energy cuts. This was necessary for this ex­periment as the stop rate for the H of the -OH group is very small (~10“ )^. Analysis of data is still in progress.The targets used for the biomedical study were contained in much thinner target hold­ers, and hence a beam of lower momentum pions would have been preferable; however, the run was still very successful. The graphite de­grader was adjusted and B^ was used to define target stops. Several different groups of target materials were examined: a) polyethyl­ene and some simple hydrides to act as stan­dards and to calibrate the experiment;b) sugar isomers to study effects seen in previous pionic X-ray measurements; c) acid anhydrides to measure changes in ir-p capture probability between molecules differing only by one hydrocarbon group in the presence of a very electrophilic group; and d) a few nitro­gen compounds.Encouraging results have been produced by preliminary analysis. The relative values for polyethylene, lithium hydride and calcium hy­dride are in good agreement with previous measurements. The results of the acid anhy­drides suggest that there is a strong connec­tion between the ir-p capture probability for a hydrogen atom and the number and bonding of the intermediate (carbon) atoms between it and the .electrophilic group, the capture probability increasing with distance. This agrees with the hydrogen bonding measurements that suggest that ir-p formation is strongly linked to local electron density. As had been suspected the measurements of the sugar isomers were all very similar, falling within 2 S.D. of their mean except for fructose at 2.8 S.D. Full analysis may resolve these re­sults but it is probable that further experi­ments will be necessary to gain statistics of better than 1% (as opposed to ~3.5%). The results for the nitrogen compounds were en­couraging but suggest that a higher pion flux may be necessary for a more comprehensive study. Further analysis on all the present data is in progress, and it is expected that further studies on M9 or M13 will be com­pleted in 1985.43Experiment 230  A comparison o f muonic molecule formation rates in HD to H2+ D 2 gases (K.A. Aniol, California State LA)In February we were given 2-1/2 days of test time on M20. We were mainly interested in the performance of the target and detection system. In particular, a measurement of the fusion neutrons from the d+d + 3He+n reaction seems to be the clearest indication of gas purity. During the test we ran four targets, l>2 , Hj+Dj, HD and H2. The H2 run served as a background run. The target was viewed by 3NE213 counters. Two of the neutron counters were 5 in. diam x 2 in. deep and the third had a diameter of 5 in. and a 5 in. depth. In addition a 5 in. x 5 in. Nal was used to look for the 5.49 MeV gammas from the d+p -*■ 3He+y fusion. The trigger for an event required a delayed decay electron from the y“ decay. The incident beam rate, 5000 to 10,000 y-/s on our 1.5 in. diam defining counter, was small enough that we could start the TDCs on a beam particle and stop on an event. If an event did not occur within 10 ys of the beam parti­cle an automatic clear signal was sent to reset the CAMAC modules. Gamma ray or elec­tron triggers could be rejected for the neu­tron event definition in hardware using the standard zero crossing of the bipolar pulse in the NE213. This greatly reduced spurious triggers due to background gamma rays. In this experiment a simple threshold cutoff is not adequate to suppress the background of gamma rays. The 2.45 MeV neutrons from the d+d fusion deposit less than 700 keV equival­ent electron energy in NE213.The combined geometrical and detection effi­ciency for the two thinner neutron counters was 0.6% and 0.7% and it was 1.2% for the larger counter. The neutron background is due to either the primary proton beam or to y~ in the experimental area. The proton beam background was measured in the identical experimental configuration during the time in between y“ beam particles. Neutron background due to y- in the experimental area was deter­mined by running the H2 target.One of the first requirements of the test run was to see how pure the gas could be made that was delivered to the target vessel. In Fig. 46 we present the spectra of energy de­posited by neutrons in the NE213 counters for two different pure D2 runs. In Fig. 46(a) we used the D2 straight from the bottle supplied by the manufacturer at 99.999% pure. The gas pressure was 720 psi. The neutron yieldis defined asHere Nn is the number of neutrons with 200 < En < 700 keV, Cjj is the deuterium concentra­tion, Np is the number of incident muons, and P is the pressure in psi. Note that there is an arbitrary normalization. The Yn quoted here are not absolute yields.The yield Yn in Fig. 46(a) is (0.7±0.1)xl0-8. In Fig. 46(b) we first passed the D2 through the silica gel U-tube in the LN2 bath. With this processing the yield jumped to (3.910.3) xlO-8, that is, a factor of 5.6 greater neu­tron flux. Thus, even at the parts per million level the use of the silica gel cold trap made an enormous difference in the puri­ty of the target gas. For all other runs [other than that in Fig. 46(a)] the silica gel trap was used.In Fig. 47 we present similar spectra for a comparison of H2+D2 and HD. In Fig. 47(a) we present Hj+D,. The yield in this case was (3.0±0.2)xl0"^ . In Fig. 47(b) the comparison run with HD produced a yield of only (1.4+0.3)xlO-8. The HD target in this case was 84%HD, 16% H2+D2. A preliminary value for theratio of yields is thusYn(H +D )preliminary    —  = 2.711.0Yn (HD)at room temperature, corrected for the D2 present in the HD target. Unfortunately this value must remain preliminary until an inde­pendent chemical analysis of our samples shows that the targets were pure to the ppm level. However, the performance of the sili­ca gel cold trap both in the manufacturing stage and in the D2 tests in Fig. 46 is promising.The gamma-ray spectra suffer from poor sta­tistics and we do not present them here. In our future runs we will have more HD (to get to 750 psi pressure) and better gamma, elec­tron and neutron detector coverage.In August we had another few days running. This run was largely developmental, although we were able to obtain new information too. We looked at muon catalyzed fusion in pure D2 (50 atm) up to 200°C. Previously the forma­tion rate dy+d -*• dyd had been measured only up to 100°C. Preliminary results indicate that the rate at 200°C is approximately 10%44En(e.e.) (keV) En (e.e. ! keV)Fig. 46. Fusion neutron spectra from muon cata­lysed fusion of d+d 3He+n. The counters are NE213 scintillators. In (a) we present the spec­trum from D2 gas with 10 ppm contamination. In (b) we present the spectrum from D2 gas which has first passed through the silica gel cold trap. Note the large difference in fusion neutron yield.160-£  120 'OmNCO8 0 -3  40OO0-20a) -8 b)Yn = (3 .0 -  0 .2 )x 10 H2 + D2 (50%  + 50% ) 4 0 0  PSIIT120-8 0 -4 0 -2 0 0  4 0 0  6 0 0  8 0 0En (e.e.) (keV)-20 H - 4 0Yn= (1.4 £ 0.3) XI0 8 HO (84% ) 370 PSI2 0 0  4 0 0  6 0 0  8 0 0En(e.e.) (keV)Fig. 47. (a) Same as 46(b) except the gas mix is 50%-50% H2,D2. (b) Same as 46(b) except the gasmix is 85% HD, 16% H2 D2. Note the large diffe­rence fusion neutron yield between 47(a) 47(b).andFig. 48. Gamma-ray spectra from the August 1984 run. (a) y 0 + background from d+pjHe+Yo* (b) background spectra.Fig. 48. We also developed a much bet­ter technique for monitoring neutron and gamma-ray background. Now the background is measured during the experiment and in such a way that there is no ambiguity in subtracting the background.higher than at 100°C. We are currently work­ing on a Monte-Carlo simulation of the exper­iment including the stopping distribution of the muons, and neutron, gamma ray and elec­tron detection efficiency. This is to deter­mine the absolute rate of the dp+d dpd formation rate.We had an improved solid angle (bigger) cov­erage over our run in February. This, com­bined with the higher target pressure, en­abled us to see the fusion gamma ray from d+p * 3He+Yo the H2+d2 mixture as shown inAnother improvement in the measurement technique was the introduction of a xenon-spiked H2 target for ranging the muon beam into the target. We found that by using the high-energy xenon X-rays we could run in singles mode and set the channel for the correct muon momentum in a few hours time. We used less than about 0.1% xenon in the H 2 target and thus our range measurement was applicable to the pure hydrogen runs.The efficiency of our impurity trapping sys­tem was tested to the limit in this run. A diaphragm on the compressor ruptured and in­jected oil into the gas manifold. It took us several days to clean out the gas manifold and because of this we were unable to run the45HD again. After the clean-up we had no dif­ficulty in observing either the neutrons from d+d -*■ 3He+n or the photons from p+d 3He+y. In fact, the figure is from data taken after the clean-up.Experiment 232  Giant muon Knight shifts (D.LI. Williams, UBC)TF-ySR with transversely polarized surface muonsA persistent problem with all high transverse field (TF) muon spin precession experiments has been the background signal from muons stopping in materials other than the sample. In other pSR techniques such as zero field (ZF) pSR, this problem is solved by using 4.1 MeV surface muons rather than ~30-40 MeV backward muons from pion decay in flight, since surface muons stop in less than 40 mg/cm2 and can be cleanly collimated to a small (~1 cm2) spot. Unfortunately, since these muons are produced with their spins antiparallel to their momentum, injecting them into a strong TF is difficult, as it is also perpendicular to their (small) momenta, and they tend to be deflected and miss the target; consequently, until recently we had to cope with background signals. Now, using M20's new dc separator we can rotate the spins of a surface muon beam by 90° (simulta­neously removing all positron contamination from the beam); the transversely polarized beam can then be injected into a strong axial TF without deflection.We have now applied this technique to mea­surements of the muon Knight shift K in small samples of SbSn alloys. We observed negligi­ble backgrounds due to muons stopping else­where; as a result it was possible to clearly identify muon trapping at unique impurity sites and to distinguish K at the trap site from that in the host.Muon diffusion and trapping in SbSn alloysZero field (ZF) experiments were performed last year on pure Sb crystals, showing clear effects of muon dynamics over the entire T range from 6 K to 200 K; the muon hop rate was rather slow at the lowest temperatures, began to speed up around 40 K, and became very fast above about 60 K. At temperatures above about 100 K the muons began to trap at unknown impurity sites.Similar measurements of the ZF muon relaxa­tion function G(t) have now been made for the SbSn(0.06%) samples. The low-T behaviour is essentially unchanged; the muon hop rate in Sb levels off below about 40 K to a constant finite value. Above 40 K the hop rate begins to increase dramatically, as expected for a thermally activated process.We turn now to our latest transverse field(TF) results. In the high quality data taken with transversely polarized surface muons (see above) we detected only one frequency in the Cu calibrations runs or in the low-T SbSn runs; yet at about 60 K in both the SbSn(0.03%) and the SbSn(0.06%) samples a second signal (at a lower frequency) begins to ap­pear, as shown in Fig. 49(a).As the temperature rises further the ampli­tude (asymmetry) of the trap site grows atthe expense of the host site in both samples, as can be seen from Fig. 49(b). This is exactly what one would expect to see when the muon is diffusing through the host lattice to find a trap site, as implied by the nomencla­ture used. As T rises the diffusion speeds up (presumably in an Arrhenius activation curve) and the muons reach the traps more quickly; this results in less phase incoher­ence at the time of transition and a larger- amplitude trap site signal.This picture is further borne out by the behaviour of the relaxation rate A of the host site signal [circles and squares in Fig. 49(c)], which decreases gradually at low T as diffusion causes motional narrowing of the nuclear dipolar relaxation but begins to increase rapidly at 60-70 K as trappingsets in. Meanwhile the muons in the trap sites relax at the same rate as those in the host sites would in the absence of trapping. Since A begins to change rather abruptly as the activation temperature is reached, the trapping rate (which is equal to A in the trapping regime) speeds up at almost the same temperature in both samples. However, one can see from Fig. 49(b,c) that both the re­laxation rate in the host site and the ampli­tude in the trap site signal are roughly twice as large in the 0.06% sample as in the0.03% sample at the same temperature, which is as expected if the Sn impurities are the trapping centres and the crystals are other­wise identical. A simple mathematical ver­sion of this model confirms these qualitative expectations.We are therefore safe in our surmise that the low-T signal is indeed due to muons in the460 10 20  3 0  4 0  5 0  60  70  8 0  90Tem perature (K)Fig. 49. (a) Knight shifts; (b) muon preces­sion signal amplitudes; (c) muon relaxation rates in SbSn crystals as a function of tem­perature. Circles and diamonds: original(low temperature) host site and high-T trap site signals, respectively, for the SbSn (0.03%) sample with H I c; squares and tri- angle^: host and trap sites for SbSn (0.06%) with H at 45° to c.bulk host material, where any change of K(host) must be due to genuine band structure effects; this is particularly important inview of the surprising increase of K when0.03% or 0.06% Sn is alloyed in pure Sb. (Con­centrations of >0.1% Sn drastically reduce K in Sb.) In addition, we have new information about K in specific trap sites, presumably adjacent to a Sn impurity. Here K(trap) is drastically reduced relative to K(host) and also seems (if we presume that there is neg­ligible difference between 0.03% and 0.06% Sn for the same orientation) to have an aniso­tropy opposite to that of the host ^ite,1.e., smaller (as opposed to larger for H I c than for H 1 c or at 45° to c.Muon Knight shifts in GrafoilA brief experiment at SIN in 1982 showed that a graphite crystal and a stack of Grafoil sheets (exfoliated pyrolitic graphite) both give the p+ a large (300-500 ppm) and aniso­tropic Knight shift at room temperature. Even in antimony the Knight shift decreases with T and becomes quite small (~50 ppm) by 300 K. Thus we were led to expect rather spectacular effects in graphite (or Grafoil, which is locally a good approximation to an aligned graphite crystal) as the temperature was lowered. This has turned out not to be the case.A short test run of a few hours showed that the muon Knight shift in Grafoil increases only gradually (by about 15%) as T is lowered from 300 K to 10 K. This is particularly surprising in view of recent results from the Tokyo group which show evidence for muonium formation in intercalated graphite samples at low T. Evidently the electronic environment of the p+ is dramatically affected by the intercalants - which in turn indicates that the p+ is probably in the same planes with the intercalants in those systems.Experiment 239  Muon spin relaxation in random spin systems (Y.J. Uemura, Brookhaven)Experiment 239 is proposed for the study of spin fluctuation and polarization of random spin systems with the use of the muon spin relaxation technique. Major effort on recent research has been put on spin glass systems. We started to apply zero- and longitudinal- field pSR (ZF- and LF-pSR) methods to CuMn and AuFe spin glasses in 1980 at TRIUMF (Expt. 71), and the results for ~1 at.% spe­cimens were reported previously [Uemura et al., Phys. Rev. Lett 45^ , 583 (1980); Hyp.471 6 0yju_oQ<cr14020<h-(i)/z.+ in CuMn5  a t .%4 3  a t .%*+ 1I at.%"'".IJ g ( la t .% )  Tg ( 3 a t .% )  Tg ( 5 a t .% )10 2 0  3 0TEMPERATURE (K)Fig. 50. Static random field distribution width as a function of temperature for several CuMn spin glasses.Int. 8_, 739 (1981); Physica 109&110B, 1915(1982)]. In the recent runs of Expt. 239 we have performed the ZF-ySR measurements on 5 and 3 at.% CuMn. The cut-off pieces of these specimens were already examined by neutron- spin-echo (NSE) and ac-susceptibility (Xac) measurements at Grenoble, and we have tried to compare and combine the results from ZF-ySR with those from NSE and Xac" Dr. A.P. Murani of Institute Laue Langevin (Grenoble) brought the specimens and participated in the pSR measurements at TRIUMF.From the muon spin relaxation function ob­served at t = 4 ns - 10 vs we deduced the dy­namic depolarization rate X<j of muon spins and the average amplitude as of static random local fields at the muon sites, as shown in Fig. 50. The rapid increase of X<j around the susceptibility-cusp temperature Tg is due to the dynamic slowing down of Mn spin fluctua­tions. The average correlation time t c  of Mn moments was determined from X^ and is shown in Fig. 51 for several different specimens studied at TRIUMF. The amplitude as of sta­tic random fields attains finite value only below Tg and increases with decreasing tem­perature towards the full amplitude a0 at T = 0. The quantity (as/a0) is proportional to the static spin polarization of Mn moments measured at the time window t = l/^s ~ 10~7 s.In Fig. 52 we compared (ag/aQ)2 of ZF-ySR (closed circles) with the static order param­eter of Mn moments determined by ac-suscepti- bility (open triangles) with the same speci­mens at the time window t = 50 ms. The good agreement of the results from ZF-ySR and Xac indicates that the polarization of Mn spinsbelow Tg persists from t = 10 7 s to t 2 ln_1 s. The Mn-spin correlation function e(t)10 'was determined accurately for 5% CuMn by NSE and Xzc as shown in Fig. 53. We plotted the results of ZF-ySR obtained for the same spec­imen, with the closed triangles for the dy­namic correlation and with the closed circles for the static order parameter (as/a0)2. The10-810 9 E-1010crcroZFyi-SR• 5 at.% CuMn•  3• 1.1* 1.4 at.% AuFe a 1.0% "NSE o  5 at.% Cu Mn10•12 I I I I 1 I 1 I I I I I Li-LLlJ0.01 0.1Fig. 51. Dynamic y+ relaxation rate as afunction of temperature for CuMn (3 at.% and5 at.%).48•  ZF-^SR (as / a 0 f oc-Sus (I - V x  )  S-K Model 0 1 T)C urie1.00 .50.0\ vAuFe 1.0 at.% 'x jU i.0 .0  0 .5T /  Tn1.0 0 .0  0 .5T / T a1.0 0.0T /  TaFig. 52. Local field correlation time as a function of reduced temperature for for a variety of spin glasses, from ySR and NSE (neutron spin echo) experiments.£  1.0o N eutron  Spin Echo Z F -^ S R o ac -S u scep t.0 - 10- 0 - 5 -5 -  '±20 K 5K  ^ CuMn (5at.% ,Tg = 27.4K) O-0-O-O- + --,6.K_X  X  Xj  V  I o  i . 0 ^ 0  ^ _| u« 0 0 -20 K26 K-6- o -2(sec)Fig. 53. Comparison of static order parameter temperature dependences measured by ZF-pSR, ac-susceptibility and neutron spin echo techniques.results of NSE, ZFySR and Xac connect nicely, and demonstrate that Mn spins slow down rapidly above Tg and possess the time-per­sisting static polarization below Tg. The The present muon experiment has provided im­portant information in the time region inac­cessible for NSE and Xac’The comparison of (ag/a0)2 with the order parameter measured by xac and by NSE in Figs. 52 and 53 also enables us to study the homogeneity of the static polarization among different Mn moments below Tg. The static field as is proportional to the Mn moment S and its concentration (see Fig. 50). There­fore, ZF-ySR is probing the linear average of static polarization (S^)^. On the other hand Xac and NSE measures (S2).^. The good agree­ment in Figs. 52 and 53, obtained by [Si)jJ2 = (S?) = (s|)^» indicates that the frozen spins share rather homogeneous amplitude of static spin polarization below Tg. These re­sults have been published [Uemura et al., Phys. Rev B 30, 1606 (1984), B ^  (in press)]Experiment 241 Temperature dependence o f muonium reaction rates (Y.C. Jean, Missouri; D.C. Walker, UBC)The 14 shifts on M20B in 1984 were extremely successful and showed that muonium is obser­vable in liquid and solid methane (45-110 K) and in liquid and solid nitrogen (45-77 K). Furthermore, on addition of 3 x 10-4 M ethene to each of these, the observed muonium relax­ation enabled tentative reaction rate con­stants to be assigned in all four phases. An activation energy of ~4 kj mole-1 was evident for the ethene reaction in liquid methane (90-110 K).The muonium and diamagnetic muon yields are themselves of considerable interest in these media. Figure 54 shows the temperature de­pendence of the diamagnetic muon signal in methane. A sharp activation-controlled dis­continuity occurs in the solid phase, but not at a known phase transition. Furthermore, the liquid phase fractional yields of muonium are ~0.2, so they differ markedly from liquid Ar and from gaseous CH^ and N2- This has bearing on the role of hot atom reactions, supporting the view that they are much more important in liquids than in gases, but only in molecular systems.49Temperature ( toFig. 54. Fractional diamagnetic muon polari­zation (Pp) in liquid and solid methane, as a function of temperature.Experiment 244  The magnetic superconductor YgCo7 (E.J. Ansaldo, Saskatchewan)Among the materials exhibiting both magnetic and superconducting properties, YgCo7 is unique in that the transition to a (ferro) magnetically ordered state (~6 K) occurs at a higher temperature than the superconducting transition (Ts=2.7 K). The transitions are broad and the two properties may coexist over a wide temperature range, up to 5 K, but the detailed nature of the coexistence is not clear in this so-called 'hybrid' state. From a variety of experiments, the magnetism ap­pears to be due predominantly to itinerant electrons. For such situations a simple mean field theory explains the lower T onset of superconductivity with subsequent quenching of the magnetic ordering below Ts. Part of the data for zero field pSR obtained at TRIUMF is displayed in Fig. 55. Firstly, above 6K (paramagnetic region) the relaxation function is due to the static Co nuclear di­poles with no evidence for depolarization due to scattering from itinerant electrons. Below 6 K, in the hybrid state a second ('fast') signal develops, reflecting the fraction of muons directly experiencing the effects of the magnetic ordering. The most striking feature of the fast signal is that it grows steadily even into the superconducting re­gion, in contradiction with the simple model. This behaviour is moreover unlike that of any other magnetic material studied so far by ySR. In terms of internal fields the fast signal corresponds to comparatively smallFig. 55. Zero field muon spin relaxation ineffective fields at the muon (<100 Oe). The relaxation of the fast signal is likely due to large inhomogeneities in the field or impurity trapping effects. In general, our results are consistent with the formation of magnetic domains in the hybrid state whichpersist and are expected to become smaller than the coherence length in the supercon­ducting region below Ts. Such an effect could be labelled crypto-itinerant-ferromag- netism in analogy to crypto-ferromagnetism in the case of localized moments magnetism.Analysis of zero field ySR data of ErRh1+Blt has advanced to the point of deducing thefluctuation rate of the local field at themuon (proportional to the fluctuation rate of the erbium magnetic moment) as a function of temperature for this material (Fig. 56) be­tween 4 K and 200 K, in a 2-muon-site model. The erbium moments (10 yB free ion) are so large that they relax the muon spin polariza­tion in less than 10 ns (our equipment'sTemperature (K)Fig. 56. Fluctuation rate of local field at the muon in ErRh^B^.50initial dead time) below about 50 K, so that only a reduced signal is observed. Nonethe­less, no dramatic effect is seen in crossing the superconducting transition temperature (8.7 K). The break in the smooth rise of fluctuation rate with temperature apparent in Fig. 56 between 155 and 200 K is probably an indication of the onset of muon hopping from one site to the other. The least squares fit model does not yet take this possibility into account so the points above 155 K only show the general trend. The deduced fluctuation rate is generally consistent with mechanisms of rare earth fluctuation expected to be operating in this material: a temperature-independent rate (at the lower temperatures of Fig. 56) due to the magnetic exchange interaction that causes magnetic ordering at lower temperature (0.7 K) plus a temperature- dependent rate due to spin lattice fluctua­tion among the f-electron states of the erbi­um ion, which are split in energy by the crystalline electric field at the rare earth site. Construction of a detailed theoretical model of erbium fluctuation, to fit to the observed rate, is now in progress.Experiment 245  Muon spin rotation studies o f platinum catalysts (R.F. Marzke, Arizona State)Two studies of platinum catalysts were con­ducted during this year, one in June and the other in December. The first study was of supported catalysts, the support used being the Cabosil silica studied extensively by Harshman et al. in Expt. 191 (now concluded) and by Keitel et al. in Expt. 288. The cata­lysts were made at Arizona State University and characterized by standard gas adsorption techniques at Stanford University. Platinum loadings varied from 0% to 1.0%, and the tem­perature ranged from 5 to 85 K. Interpreta­tion of the observed dependences of the muonium relaxation rates upon temperature and platinum loading showed the following: 1) The relaxation rate peak seen in the range 20- 40 K in the unloaded support is removed by the presence of even a very small amount of Pt (as little as 0.001% by weight). 2) Muon­ium reacts chemically with the primarily oxygen-covered Pt surface at the rate of 3.5±0.15 inverse microseconds. The former effect is explained by the reduction of the trap responsible for the 20-40 K peak by atomic hydrogen, which is generated during the dihydrogen reduction step in the produc­tion of Pt-loaded catalysts (but not in un­loaded ones). The latter effect representsthe first measurement of a rate constant for a heterogeneously catalyzed reaction involv­ing muonium, and it establishes a bound for the corresponding protium reaction rate.The second study of Pt catalysts conducted during 1984 centred on unsupported, finely divided material known generically as plati­num black. Data from these runs are now be­ing analysed, and so results are not yet available. Preliminary fits made during the experiments show substantial effects of gas adsorption upon the muon relaxation rate, however, in agreement with earlier studies conducted in Expt. 191.Experiment 260  The reaction o f muonium with hydrogen peroxide in water (P Percival, Simon Fraser)The reaction of muonium with hydrogen perox­ide is being studied to clarify the mechanism of the corresponding H atom reaction. Two pathways are known for the gas phase reactionH + H202 ■* H20 + *0H (1)H + H202 ■+ H2 + •00H , (2)but there is ambiguity as to which dominates in aqueous solution.Last year we determined the rate constant for Mu + H2C>2 at natural pH to be kjq = (1.8±0.2) xlO9 M-1 s-1. Furthermore, we conducted a residual polarization study of concentrated H202 solutions which showed that the muon product of the reaction is diamagnetic. A preliminary determination of the rate con­stant for Mu + D202 at natural pD suggested that it was lower than that for H202 by a factor of two. This seemed to point to the H abstraction reaction (2). However, this year we repeated the measurements under more care­fully controlled conditions and we find no isotope effect.Also last year we discovered that the effec­tive rate constant for Mu + hydrogen peroxide is pH dependent. This is particularly inter­esting since there is no equivalent data for the II atom reaction. We followed up the H202 study with one using D202. The data obtained to date are shown in Fig. 57. Analysis of both sets of data was carried out under the assumption of competitive parallel reactions of Mu with H202 and H02 (D202 and D02). The results confirm the lack of isotope ef­fect in the reaction of the neutral molecule,51| D20 2 ]/m MFig. 57. pD dependence of muonium decay rates in D202 solutions.but suggest that there may be an effect for the basic form:k(Mu+H202)k(Mu+H02)k(Mu+D202)k(Mu+D02)(1.54 ± 0.08) x 109 M-1 s_1 (4.9 ± 0.6) x 109 M'1(1.52 ± 0.02) (3.3 ± 0.4)~" s' 9 M- 1x 103 M' x 109 M-1 s'1hexane, suggesting a relatively long-lived neutral precursor. (iii) Addition of CCi.^  at ~1 M did not prevent radical formation, nor seriously disturb the selectivity, so the precursor is not readily scavengable by car- bontetrachloride (whatever that now means! - see Expt. 157 report).Experiment 275Search for superdiffusion in stressed vanadium (J. Bailey, DESY)diumExperiment 262Muonium-radicai formation mechanism(D.C. Walker, UBC)These experiments were the first to utilize a new ySR clock on M20 and, unfortunately, the clock gave a spurious 320 MHz signal that has rendered the quantitative enumeration of cyclohexadienyl radicals doubtful. So the results from the 10 shifts in 1984 have only qualitative significance. In the high field limit the pair of precession frequencies of a free radical containing y+ sum to the muon- electron hyperfine coupling constant, with benzene ring radicals lying at 490-510 MHz and aliphatic side chains at 200-300 MHz.Three qualitative results were obtained, (i) Allyl benzene showed both cyclohexadienyl and side-chain radicals, suggesting that the single radical in styrene stems from intra­molecular relaxation rather than selectivity in the primary kinetics of the radical pre­cursor. (ii) The 80% benzene-20% styrene mixture that showed equal numbers of radicals from benzene and styrene [see 1982 Annual Report, Expt. 157] gave the same free radical distribution when diluted to ~5 M by cyclo-Recent work by Suzuki et al. on H and D dif­fusion in V single crystals has shown a re­producible increase of the diffusion constant by nearly two orders of magnitude upon appli­cation of tension along a certain crystalline axis. Experiment 275 to investigate the analogous effect for muons in V was approved in December 1983, and preliminary data were taken early in 1984. No evidence for such a dramatic enhancement of the diffusion rate was found, indicating that the phenomenon observed for H and D is associated with H-H or D-D interactions (those measurements were made with high concentrations of H and D).Experiment 286Quantum diffusion of muons and muonium(J. Brewer, UBC)This newly approved experiment has received only a few shifts of beam, which were devoted to the development of a new ySR technique which holds great promise both in the eluci­dation of muon location/motion in crystals and in the use of the muon as a probe of host energy levels. This technique was simultane­ously conceived by Dr. S.R. Kreitzman and by Prof. A. Abragam, whose timely suggestions are greatly appreciated.Level-crossing resonance (LCR) ySRWe have observed level-crossing resonance muon spin relaxation (LCR-ySR) in Cu at low temperatures by matching the Zeeman splitting hvg of the muon in the applied longitudinal field (LF) to the level splitting hvq of the spin-3/2 Cu nuclei in the electric field gradient (EFG) caused by the muon. On reso­nance, energy is exchanged between the pola­rized muon and the Cu nuclear spin degrees of freedom on a time scale characterized by the dipolar coupling between them.The potential impact of this work on ySR is enormous. First, the method permits direct52measurement of the physical constants of the microscopic Hamiltonian between the muon and its environment, rather than the phenomenolo­gical parameters often yielded by convention­al techniques. Second, the observed relaxa­tion is sensitive to the type and number of nearest neighbour (NN) impurities with spin > 1/2; one can in principle observe LCR-pSR separately with each such neighbour at a dis­tinct resonant field. Since much uSR work is circumscribed by impurity problems, any definitive information about the role of im­purities is vitally important. Third, the effect is expected to be extremely sensitive to muon motion. Indeed, for Cu the entire quadrupolar energy of the NN nuclei origin­ates from the presence of the p+ , which breaks the otherwise cubic symmetry of their environment. Thus as the muon hops it carries the LCR interaction along with it, causing a broadening of the resonance. This effect can be exploited to obtain detailed information on the still-controversial quan­tum diffusion of the p+ in Cu below 5 K, among other applications.Figure 58 shows an LCR-p+SR spectrum taken at 20 K on a Cu single crystal oriented with the <111> direction parallel to the magnetic field. Each data point represents 20 min of beam time. Despite the long relaxation times (many muon lifetimes) we still see a dramatic resonance curve with a central resonant field of 80.9 Oe. This directly confirms previous models of the nature and size of the disturb­ance caused by the p on the NN Cu nuclei.Gaussian Line +  t e r m / ( H  +  Hc2)U)3.20  40  60  80  100 120Experiment 290  Positive muon probing solitons in polyacetylene (K. Nagamine, Tokyo)Trans-polyacetylene, (CH)X or (CD)X , has at­tracted so much theoretical and experimental interests because of the proposed existence of solitons which can be ascribed to one­dimensional diffusion of unpaired electrons associated with a different phase of bond alternation. Recently it was found that the unpaired electron produced by the injected positive muon takes a one-dimensional soliton-like motion in trans-polyacetylene [Nagamine et al., Phys. Rev. Lett. 53, 1763 (1984)]. The aim of the present experiment conducted on M20 and partly at M15 was to see the dynamical behaviour of the muon-induced soliton through the p+ longitudinal relaxa­tion at various temperatures (from 3.8 K. to 300 K) and applied longitudinal fields (from 0 G to 2.0 kG).At all the measured temperatures we found that the p+ relaxation rate changes smoothly against applied field H following basically an (H)-1/2 law consistent with the picture of p+ relaxation due to one-dimensionally moving unpaired spin. However, at low field it was found that there is a clear temperature- dependent change in the shape of u relaxa­tion functions. As a rather qualitative rep­resentation of the effect the asymmetry and the relaxation rate in zero-field relaxation are presented in Fig. 59(a) and (b), respec­tively, where we have assumed a simple expo­nential relaxation function. It can be seen that a significant change exists at around 150 K, which might be related to a change of dynamics of the muon-induced soliton, like an occurrence of confinement of the soliton at low temperature. In order to deduce a clearer conclusion sophisticated data analysis is required based upon the model of p+ relaxa­tion functions reflecting the soliton-dynam- ics with various corrections such as broaden­ing due to surrounding proton dipole moments, etc. Work along this line is now in progress.Experiment 296  Muonium addition reaction in the gas phase (D. Garner, TRIUMF; D. Fleming, UBC)Longitudinal Field (Oe)Fig. 58. p+SR level crossing resonance with Cu nuclear quadrupole splittings in Cu crystal at 20 K with <111> axis parallel to longitudinal applied field.As part of Expt. 147 we measured the reaction rate of Mu with ethylene (C^^), which estab­lished that Mu reacted much faster than the H atom at temperatures T < 200 K; this pro­nounced reactivity demonstrates the sensitiv­ity of the much lighter Mu atom to quantum53100 200 T E M P E R A T U R E100 200 T E M P E R A T U R E [CO] (10* molecule cm1 )Fig. 59. (a) The temperature dependence ofthe asymmetry in zero-field longitudinal re­laxation functions obtained by a preliminary fit with an exponential relaxation function, (b) Same for the relaxation rates.tunnelling [see 1982 Annual Report; Garner et al., Chem. Phys. Lett., to be submitted]. More recent work with C2D[t has indicated a slight (secondary) isotope effect not seen in the corresponding H(D) atom data [Sugawara et al., Chem. Phys. Lett. !?>_> 259 (1981)]. These experiments prompted a more general study of Mu addition reactions, the subject of Expt. 296.To date we have obtained preliminary data at 300 K for the reaction of Mu with C02 (none), CO, S02 (impure), allene and acetylene. The corresponding H atom reactions with these species are known, that with C02 being an abstraction reaction which accounts also for its unobservably slow Mu reactivity. Typical results for the Mu + CO reaction in the form of relaxation rate X vs. [CO] are shown in Fig. 60; the slope gives the bimolecular rate constant k » 4.2*10-13 cc s_1. This is about 10 x faster than the corresponding H atom re­action, indicative again of the significance of quantum tunnelling in Mu reactivity. Further work is in progress, including a study of the T dependence of these reactions.In collaboration with Dr. Emil Roduner from SIN a fairly complete set of data has been obtained on the temperature dependence of theFig. 60. The Mu relaxation rate X vs. [CO] concentration at 300 K; the slope gives the bimolecular rate constant, kj^ u = 4.2x10“13 cco  1 _reactivity of Mu with dimethylbutadiene (DMBD) and with benzene. [Although started under the aegis of Expt. 147 (in February 1984) it is more naturally included here as part of Expt. 296.] The Arrhenius plot for the T range ~300-500 K is shown in Fig. 61,Coo1000/Tem perature (K '1)Fig. 61. Arrhenius plot for Mu + DMBD and Mu + benzene in comparison with H + butadiene and H + benzene.54Table IX. Mu and H Arrhenius parameters for DMBD and benzene.Reaction k300(10 12cc s A(10 12cc s 1) Ea(kcal mol 1)Mu + DMBD 70.7 ± 2.5 178 ± 22 0.55 ± 0.07H + butadiene 7.8 68 1.3Mu + benzene 1.1 ± 0.2 16.2 ± 2.0 1.60 ± 0.07H + benzene 0.08 66 4.0which presents also the corresponding plots for the H atom results. These results are summarized in Table IX (H+DMBD not measured but should be similar to butadiene itself). Again at room temperature, W kH ~ iO/U moreover, the activation energies Ea for the Mu reaction are significantly less than for the corresponding H atom reactions. This is perhaps the most clear-cut indication of the dominance of quantum tunnelling in Mu reactivity, demonstrated also in our muchearlier results for the Mu + F2 reaction [Garner et al., Chem. Phys. Lett. 55_, 163(1978)]. Present experiments at SIN by Roduner are concerned with measuring the T dependence of these reactions in the liquid phase. A comparison of those results with the present gas phase ones (including more additional T points) will yield valuable information on the effects of solvation and diffusion in the most elementary of chemical reactions.55THEORETICAL 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 problems relevant to a pion fac­tory and the proposed kaon factory. The re­search activities of the group cover a wide spectrum of interests, from uSR to Higgs boson production, from generating curves for the experimentalists to mathematical physics. This wide variety of interests is necessary in order to be of maximum use to TRIUMF in both the long and short terms.Currently there are three permanent staff members in the group: H.W. Fearing, B.K.Jennings and R.M. Woloshyn. A fourth, A. Picklesimer, will be joining the group next summer. J.N. Ng holds an NSERC University Research Fellowship. The research associates, some of whom are supported through NSERC grants, are C.Y. Cheung, E.D. Cooper (from September), G. Fogleman (to September), S. Godfrey (from October), M.J. Iqbal (from Sep­tember), K. Masutani (to June), O.V. Maxwell (to August), J. Thaler (from September), R.E. Turner and W. Wilcox. Long-term visitors who have been with the group are S. Gurvitz, R. Machleidt and E.A. Veit. In addition there is interaction with theorists from the four associated universities.Members of the group have taught courses at UBC and supervised graduate students. Current graduate students are G. Couture, D. Hamilton and R. Workman.In addition to their research activities the theorists are also involved in a number of the laboratory activities. These activities include involement with the Long-Range Plan­ning Committee and the Kaon Factory Steering Committee. Organization of the TRIUMF semi­nars has also been handled by the theory group. This and the summer theoretical visi­tors program have brought a large number of visiting theorists to TRIUMF including:R. Lawson G. Miller T. Mizutani M. Moravcsik A. PicklesimerA. Rinat D. Schiitte J. Sheppard A.W. Thomas G. TupperM. Turner J. Vergados R. Vinh Mau G. WalkerI. Affleck B. Campbell D. HeddleJ. Bagger L.N. Chang K. HolindeR. Barrett T. Draper P. KalyniakI. Bigi H. Esbensen A. KobosB. Blankleider R. Fiebig K. KuboderaE. Braaten T. Goldman H. LamR. Brockmann J. Greben R . LandauA. Brown H. Haber J. LawSpecific research activities undertaken dur­ing the last year are outlined below.Nucleon-nucleon potential and applications to finite nucleiMass difference effects in the NN system (J. Thaler)The model dependence of the difference be­tween the hadronic pp and the nn s-wave scat­tering has been studied in terms of off-shell effects of the hadronic NN interaction. By comparison with experimental results we ob­tained a strong constraint for the corre­sponding half-shell function. Charge-symmetry breaking effects play an important role in the description of the data by realistic NN potentials like the Paris potential.Charge independence breaking in a meson theory of NN interactions (C. Y. Cheung; R. Machleidt, Bonn and TRIUMF)A fundamental measure of the charge depend­ence of NN forces is the difference between the pp (or nn) and np 1SQ scattering lengths. Experimentally, one finds l/2(app+ann) - anp “ 5.5 fm [Dumbrajs et al., 1982 Data Compila­tion, Nucl. Phys. B216, 277 (1983)], whichhas never been fully explained despite much theoretical effort.In this work we try to address this problem within the framework of a consistent meson model of NN interactions [Machleidt and Holinde, Proc. Few Body X, Karlsruhe, 1983, Vol. II; Machleidt, Proc. Quarks and nuclear structure, Bad Honnef, Germany 1983, Lecture Notes in Physics 197, 352 (1984)]. The model contains the well-known one-meson-exchange contributions of it, o and <o. In addition, it also has the 2ir-exchange contributions with virtual isobar excitations and direct tttt in­teractions taken into account. This part of the model is in agreement with dispersion theoretic results as well as the empirical NN phase shifts. The model also includes ir­reducible diagrams of irp exchange which are crucial for the description of low partial56waves. Furthermore, important irreducible 3tt and 4n contributions are taken into account effectively by ircr and iru> exchanges.Within this meson-exchange picture charge independence breaking may arise from the mass differences between the charge states of it, p, N and A. In addition, the isospin depend­ence of meson-baryon coupling constants and photon exchange can also contribute.In this summary we concentrate only on the charge-dependent effects due to the pion mass difference. The average nucleon mass M = 938.926 MeV and A mass M^ = 1232 MeV are used throughout _the calculation. We first fit the quantity a = l/2(app+ann) = -18.2 fm, and then successively examine the effect of ir-mass difference in various sectors of the NN interaction model described above. The results are summarized in Table X, where 6a = a - anp and the intermediate baryon states in the two-boson-exchange contribu­tions are indicated in parentheses. We find that the largest effect occurs in OPEP. Note that individual contributions in TPEP can be large; however, they partially cancel and the resultant effect is much smaller than that in OPEP. Also, without exception, we find par­tial cancellation between contributions from 'box' and ’crossed-box' diagrams in TPEP. The effect of ir-mass difference in TPEP (NN) has also been calculated by Ericson and Miller [Phys. Lett. 132B, 32 (1983)]. They found 6a = 0.88 fm, which is much larger than our result of 0.18 fm. The reason is most likely due to the fact that we have excluded anti­nucleon intermediate states in TPEP (NN), while they are included in the model used in their work. However, in a consistent meson- exchange model of NN interactions the inclu-Table X. Various contributions to the charge dependence of the singlet scattering length length due to the pion mass difference.Potential 6a (fm)OPEP 2.80TPEP (NN) 0.18TPEP (NA) 3.25 0.85TPEP (AA) -2.58irp-exchange (NN) -0.39ira, irw-exchanges 1.18Total 4.44sion of antinucleon states in TPEP (NN) would render the medium-range attraction too large. Therefore, the effect of the ir-mass differ­ence in TPEP (NN) found in the Ericson and Miller result is probably overestimated.Finally, we concluded that the pion mass dif­ference can account for 4.44 fm of the diffe­rence between & and anp. This is about 80% of the experimental value (= 5.5 fm).The 2n- exchange potential for pseudovectorcoupling o f pions and nucleons(J. Thaler)A 2ir-exchange potential for pseudovector coupling of pions and nucleons is derived by transforming the Dirac equation into a Schrodinger-like form. The resulting TPE potential is spin dependent_and attractive in all states of the NN and NN system and de­scribes part of the intermediate range inter­action. A comparison is made with the Hamada- Johnston potential and with the TPE potential of different approaches.The 2n- exchange potential in nuclear systems(J. Thaler)The effect of a constituent quark structure of nucleons on the 2ir-exchange potential for the NN system is studied. While the assump­tion of elementary nucleons and the constitu­ent model lead to the same result for the OPE potential, an essential difference is found for the TPE contribution, especially for the central potential in spin-singlet states of the NN system. The resulting TPE potential is able to give a satisfactory description of higher partial wave phase shifts of the na and the aa system.Effective nucleus-nucleus potentials from meson exchange(J. Thaler)The effective a-a potential is probably one of the best-defined strong interaction poten­tials in nuclear physics. It allows a very simple parametrization and has a unique form in co-ordinate space. It was found for the first time that this a-a interaction can be uniquely described by 2ir and w exchange for a-a separations above 2.5 fm. The study is extended to other nucleus-nucleus systems with heavier nuclei. In most of these systems the optical potential is not unambiguously known, and the description of the long-range behaviour by 2ir and w exchange is found to reduce the ambiguity considerably.57Bremsstrahlung and off-shell processesPotential model calculations of proton-proton bremsstrahlung (H. W. Fearing, Ft. Workman)During the past year there has been major progress on our program of theoretical calcu­lations of ppy, and the first stage is now essentially complete. Recall that the aim of these calculations was to develop a 'good' and 'modern' potential model calculation for comparison with the data and with the older on-shell soft photon approximation calcula­tions. By 'good' was meant a calculation which contains a number of effects such as relativistic spin corrections, Coulomb ef­fects, one-pion-exchange contributions to the higher partial waves, etc. that had not pre­viously been included simultaneously, and by 'modern' was meant a calculation using one of the modern theoretically based potentials such as the Paris or Bonn potential.We now have such a calculation working for both cross sections and analysing powers us­ing both Paris potential and a Reid soft core potential, extended to higher partial waves. The method used is to solve the Lippmann- Schwinger equation in momentum space for the half-off-shell nucleon-nucleon T-matrix which is then combined with propagator factors and the electromagnetic vertex to get the ppY am­plitude. Such a calculation thus includes both off-shell effects and terms of all ord­ers in the photon momentum k in distinction to the soft photon approximation which re­quires purely on-shell information and in­cludes only the first two orders in k.For the Paris potential the results for both cross section and analysing powers are dis­tinctly different from those obtained via the soft photon approximation, and the experiment at 280 MeV now under way at TRIUMF should easily distinguish between potential and soft photon predictions. What is somewhat surpris­ing, however, is that the Paris potential re­sults are quite similar to older calculations with the original Reid soft core of other phenomenological potentials, and thus still in rather poorer agreement with the cross- section data from the old TRIUMF experiment than are the results from the soft photon approximation.To understand this similarity we have also calculated the half-off-shell extension func­tion for both Paris and extended Reid poten­tials and found them to be very similar. Thus in retrospect the similarity of all the po­tential model results does not indicate a lack of sensitivity in ppy to off-shell ef­fects, but instead simply reflects the fact that all potentials we have tried so far are very much the same both on and off shell.As a second stage there are several refine­ments which eventually could be included. Our calculation is gauge invariant as it stands, but there are additional gauge terms coming from nonlocalities in the potential which should be added. These have never been cal­culated before, but probably are small. We have also not yet included higher-order pieces of the double scattering terms. How­ever, if these latter corrections have the same sign as at somewhat lower energies they will increase the difference between soft photon and potential model results.Proton-proton bremsstrahlung using the Bonn potential (H. W. Fearing, R. Workman; R. Machleidt, Bonn and TRIUMF)As an extension of the previous calculation we have also calculated pp bremsstrahlung using a slightly different method and the Bonn potential as input. The starting point here was the half-off-shell nucleon-nucleon T-matrix as generated by the Bonn group. This half-shell T-matrix was then inserted in the rest of the calculation as described above.Results using the Bonn potential are in gene­ral very similar to the Paris results, for both cross section and analysing power. Al­though we have found one or two isolated kin­ematic situations when Bonn and Paris analys­ing powers are quite different, we have not yet found a situation where the experiment under way at TRIUMF is likely to be able to distinguish the two potentials. Again this similarity can be traced to the fact that the off-shell extension functions are very simi­lar to those for the Paris potential for most partial waves.Soft-photon theorems as a probe o f nonradiative amplitudes (H. W. Fearing; G. Walker, Indiana)Soft photon theorems relate radiative ampli­tudes to the corresponding nonradiative amp­litude. Thus given a complete knowledge of the nonradiative amplitude one can calculate exactly the radiative amplitude through the first two orders in the photon momentum k and express it in terms of the nonradiative amp­litude and, what is important, also the der­ivatives of the nonradiative amplitude. Nor­mally the nonradiataive amplitude is known58and this approach is used to estimate the radiative amplitude.However, for some processes the nonradiative amplitude may not be well known. Since the radiative amplitude in the soft-photon region involves both amplitudes and derivatives it provides an independent constraint, just like a spin correlation would, on the nonradiative amplitudes. Thus there might be some situa­tions where one could reverse the usual pro­cedure and use the radiative process to get additional information on the nonradiative process. We are currently looking at the feasibility of this idea with respect to pion nucleus scattering.Relativistic effects in nuclear physicsNuclear optical potential in a meson-exchange model o f nucleon-nucleon interaction (M.J. Iqbal)Dirac phenomenology has been very successful in fitting experimental data for elastic and inelastic proton scattering on different closed-shell nuclei. An interesting aspect of Dirac phenomenology is the existence of strongly attractive scalar and repulsive vec­tor (fourth component of a four-vector) po­tentials. It will be interesting if one could explain these strengths in a 'more fundamen­tal' model of nuclear forces, e.g. meson- exchange model. Recent calculations show that indeed is the case [Iqbal, preprint TRI-PP-84-111, submitted to Phys. Lett. B]. More detailed calculations including higher- order Born terms and Pauli-blocking effects in the nuclear medium are currently being completed.Comparison o f Dirac and Schrodinger calculations (E.D. Cooper; A.O. Gattone, M.H. Macfarlane, Indiana)A straightforward scheme for comparing Dirac phenomenological and Schrodinger calculations has been developed [Cooper et al., Phys. Lett. 130B, 359 (198 )]. The essence of the scheme is that 2-component operators are found which have simple forms but which when used in place of Schrodinger operators along with Schr'ddinger wave functions reproduce the results of a Dirac calculation. Several reactions were considered as a possible place to look for 'Dirac effects' by means of com­paring these 2-component Dirac operators to the usual Schrodinger ones. Most of thedifferences between the operators appeared only in the nuclear surface, and were spin dependent, but no clear Dirac signatures were found.Relativistic effects in protonium (J. Thaler)Relativistic corrections to hadronic energy shifts in protonium are discussed, and the main effect is found to be due to a 2ir-ex- change potential, which influences the medium range behaviour of the NN interaction. In view of application to heavier antiprotonic atoms also the validity of the Born approxi­mation for the hadronic interaction has been checked for various states of protonium.Derivation o f Dirac DWBA and application to proton inelastic scattering (E.D. Cooper; H.S. Sherif, R. Sawafta, J. Johansson, Alberta)It has been found that one can derive the Dirac DWBA from an adding Hamiltonian pre­scription. Such a prescription leads to dis­aster for the nucleon-nucleon case, the Breit equation [IUCF internal report, in prepara­tion] , whereas it might be all right to add the Hamiltonians for fermions and bosons.Reactions that can be described in this way include (p,ir+) and (p,p') to states which can be described as a single phonon excitation of the target nucleus. The agreement with data for all spin observables is surprisingly good for both the above reactions, indicating the assumption may be a reasonable one.Nuclear saturation in a relativistic Bruckner Hartree-Fock approach (R. Machleidt, Bonn and TRIUMF; R. Brockmann, Regensburg)In the past extensive efforts have been de­voted to the explanation of the binding ener­gy per nucleon and saturation density of nuclear matter. The starting point of most calculations is either a purely phenomeno­logical NN potential or a one-boson-exchange potential (OBEP), which fits the data of free NN scattering and the deuteron. These NN potentials are used to derive a nonrelativis- tic G-matrix. It turns out that all calcula­tions yield values for the binding energy and saturation density which are located on a narrow band (the Coester line). However, this line does not pass through the empirical saturation point of nuclear matter.59Standard nuclear matter Bruckner calculations include only diagrams up to the second order in the hole-line expansion. Therefore, there had been hope that the inclusion of higher orders may solve the nuclear matter problem. However, B.D. Day has shown that the 3- and 4-hole-line diagrams improve the saturation only insignificantly apart from little depen­dence on the detailed structure of the NN interaction applied.One attempt out of the problem is the intro­duction of additional degrees of freedom. Assuming the nuclear force being mediated by mesons it is obvious to consider mesonic de­grees of freedom in nuclear matter. Further, an explicit model for the 2ir-exchange contri­bution to the nuclear force suggests special effects due to the virtual excitation of A isobars. Detailed calculations have shown that these effects move the saturation point up within the Coester band.A new approach to nuclear matter is the in­clusion of relativistic effects, which is the object of this work. The main difference of relativistic calculations compared to the nonrelativistic ones is the explicit treat­ment of the Dirac spinor representing a nuc­leon in nuclear matter. A consequence of this is that the 'small' components of the Dirac spinors are increased. It turns out that this strengthening of the lower component within the framework of a and w exchange creates an additional saturation mechanism in nuclear matter. The effect is highly density depen­dent and leads to a saturation point which is in agreement with the empirical data.In contrast to earlier work by Shakin and co­workers we have performed the selfconsistency in nuclear matter for the single-particle wave functions and energies consistently. We also apply the pseudovector coupling of the pion in NN scattering and nuclear matter.In conclusion we may state that we have con­structed a relativistic OBEP which for the first time describes NN scattering and nucle­ar matter quantitatively without a readjust­ment of the parameters in the nuclear matter problem. This OBEP is also a good starting point for the evaluation of a T-matrix to be applied in N-nucleus calculations in the Dirac approach.Relativistic effects in nuclear physics (S. Gurvitz)There are indications that the relativistic effects may be important even in the problemsof conventional nuclear physics. We concen­trated on the effect of the Lorentz contrac­tion of the two-nucleon system which may be the dominant relativistic effect in the cal­culations of nuclear form factors at large q2 and also in the three-body and the nuclear matter calculations. In order to treat these effects in a consistent way we performed the exact three-dimensional reduction of the Bethe-Salpeter equation. We used the light cone variables, which considerably simplify the results, but also retain the covariance of the final answer.We found a new three-dimensional relativistic equation which is equivalent to the four­dimensional Bethe-Salpeter one. In the c.m. frame our equation is very similar to the usual Lippmann-Schwinger equation and both coincide in the nonrelativistic limit. We therefore identify the solution of our equa­tion in the c.m. frame with solution of the Lippmann-Schwinger equation with phenomeno­logical potentials. However, the Lorentz boost following from our equation is now uni­quely defined. Application of our result is in progress.Nuclear reactions and electromagnetic effects(p,y) reactions in a relativistic Dirac impulse approximation (H.W. Fearing, E.D. Cooper, J. Iqbal)There has been a great deal of interest re­cently in the so-called Dirac approaches to proton-nucleus interactions at medium ener­gies . In such approaches single-particle bound or scattering wave functions are deter­mined by solving the Dirac equation in the presence of a strong scalar and vector poten­tial. Such potentials are generated either phenomenologically by fitting elastic scat­tering, in To approximation from on-shell nucleon-nucleon data, or semi-microsopically from mean field theories. Processes such as inelastic proton-nucleus scattering are then described in impulse approximation by taking matrix elements using these wave functions of an operator derived from the appropriate on- shell process. This approach has been rela­tively successful in describing certain spin correlations which are not well fitted by usual Schrodinger distorted wave calcula­tions .In view of these successes it is of interest to test the approach in other processes. We have thus begun a calculation of the reaction p+A (A+l)+y. The idea is to use the same60kind of Dirac distorted waves and bound state wave functions as used in other calculations. The interaction will be derived from the on- shell NN ■+• NNy amplitude in a fashion analo­gous to that used for other processes. We work in momentum space and have taken some pains to choose variables which will make the form of the interaction as simple as possi­ble. This puts the complications into the nuclear physics which, once done, can be used for other reactions as well, such as the very similar p+A (A+l)+n reaction.So far much of the formalism has been worked out and some preliminary numerical work has been done, to explore the feasibility of a full calculation. It appears from prelimin­ary results that such a calculation is pos­sible, and so we propose to proceed with the full calculation in the coming year.The role o f the A(1232) in the (y, p) reaction (C.Y. Cheung; B.D. Keister, Carnegie-Mellon)The (Y,p) reaction at intermediate energies has been the subject of considerable theoret­ical and experimental interest. The original expectation for this reaction was to learn about the high-momentum behaviour of nucleons in nuclei, using plane wave impulse approxi­mation (PWIA). In the PWIA description of the reaction one expects the (Y,p) cross sec­tion to be much larger than the (Y,n) cross section. However, to the contrary, experi­mental data show that the two cross sections are approximately equal. This strongly sug­gests that other reaction mechanisms, such as the two-body mechanism, must be important. Recent theoretical investigations [Londergan and Nixon, Phys. Rev. C 19^ , 998 (1979); Gari and Hebach, Phys. Rep. 72, 1 (1981)] do con­firm that for Ey > 100 MeV the two-nucleon current contribution can be sizable, if not completely dominant • At Ey = 300 MeV the photon can also excite a nucleon to a A(1232). Given the large amplitude for this process one expects it to enhance the two-body cur­rent in this energy region.Despite the substantial effort invested in the (Y,p) reaction to date, our present un­derstanding of this process is not yet clear. While it is generally agreed that the two- body current gives a large contribution for photon energies of several hundred MeV, the important elements of the two-body current are still a matter of debate. For example, Londergan and Nixon have evaluated the inter- mediate-A contribution to 160(y,p)15Ng.s . an<^  found it to be dominant in the A-resonanceenergy region. However, Gari and Hebach found that one-body and nonresonant exchange currents are very important, even for Ey = 300 MeV. They also claimed that the A cur­rent does not dominate and its inclusion is not needed to explain the general features of the data. For the energy-dependent differen­tial cross section the data show a character­istic A enhancement at Ey - 300 MeV onlyfor the case of Y d  -»■ np and not for heavier nuclei. The absence of a recognizable en­hancement means that the role of the A is not obvious, and one must turn to a calculation to examine its effect.In light of the present status of both theory and experiment we have re-examined the con­tribution of the A term to the intermediateenergy (Y,p) reaction, using an approachwhich includes pion rescattering and proton distortions (see Fig. 62). Specifically, we have evaluated the magnetic Born and inter- mediate-A contribution to the ( y , p )  reaction for ^He and 160 targets. The results of our calculations agree with those of Londerganand Nixon on the point that the one-body cur­rent is negligible in the A-resonance region. However, the size of the intermediate-A con­tribution we find is comparable to that of Gari and Hebach. It is smaller than what is calculated by Londergan and Nixon, sometimes by an order of magnitude. We conclude that the A contribution is not dominant, but not negligible either.Y(c)Fig. 62. (a) Born contribution, (b) A(direct) contribution, (c) A(exchange) contribution.61More theoretical effort is needed. Most im­portantly, the two-body electric current must be evaluated accurately in a gauge-invariant way so that direct comparison can be made with the results of Gari and Hebach for the nonresonant exchange current contribution.Pion production from nuclei (M.J. Iqbal; G.E. Walker, Indiana)Understanding the (p,ir) reaction on complex nuclei has been a difficult theoretical prob­lem. There now exists a computer code to study the (p, tt ) reaction in a two-nucleon mechanism (TNM). Initial calculations for the 12C(p,ir+)13C (9+/2) reaction at proton labkinetic energies Tp = 200, 250 and 300 MeV have been very encouraging [Iqbal and Walker, Univ. of Maryland preprint 0R0 5126-233].Currently experiments are being performed at Tp = 350 MeV and there is already a theo­retical prediction for the results. These calculations have been extended to study (p,ir) reactions at very low energies. Those results will be published shortly. The code can be used to study (p,ir~) and (ir-,p) reac­tions as well.Sensitivities in the Dirac DWBA for ( p , n )  reactions (E.D. Cooper; H.S. Sherif, Alberta; G.A. Miller, Washington)The pion stripping model for (p,ir+) reactions appears to do surprisingly well for stripping into states that can be expressed as a single nucleon state added to the target nucleus. The effect of using different pion distorting potentials on the calculation has been exam­ined [Cooper et al., Phys. Rev. C 30, 232(1984)]. The effects are not large provided the pion potentials are parametrized in such a way as to include the angle transformation and the LLEE effect.The effect of the pion form factor at the NNir vertex has also been looked at; it lowers the cross sections uniformly by a factor of two. A new project has begun to examine the ef­fects of not allowing any contributions from the NNtt vertex when the pion and the nucleon from which it is produced never come closer than 1 fm. This enables one to ask how sens­itive the stripping model predictions are to possible deviations from the meson-baryon picture at short distances. The conclusion at the present time is that there is a tre­mendous sensitivity to this short-ranged behaviour.Inclusive nuclear reactions with large momentum transfers (S. Gurvitz; A. Rinat, Weizmann; J. Tjon, Utrecht; S. Wallace, Maryland)It was demonstrated that the large angle in­clusive nuclear reactions ttA ir'X, pA p'X in the region of small energy transfers are going through the single-scattering mechanism and can be used for the extraction of the nuclear response function just in the same way as electron inclusive reactions eA e'X [Gurvitz, TRI-PP-84-85]. Similar investiga­tion of pA ttX reactions is currently in progress.The use of hadron probes for extraction of nuclear information allows one to obtain in­formation on the nuclear response which is not available from the electron data. How­ever, the question how to find the nucleon momentum distribution from the nuclear re­sponse function still remains a main problem. We found that the final-state interaction (FSI) of a disintegrated target plays a very important role. It can be partially taken into account if one uses y-scaling for the extraction of one-nucleon momentum distribu­tion. However, we found that the y-scaling is in disagreement with numerical results of model calculations. The different treatments of FSI are currently in progress.Mesonic and antiprotonic atomsThe low- and medium-energy K  p  interaction (J. Thaler)Results for the strong interaction effect in the ground state of kaonic hydrogen are pre­sented. The calculations are based on a phe­nomenological separable_potential model which describes low-energy KN data. The results disagree with the available atomic data and cannot describe the experimental discrepancy between scattering and bound-state data. By application of model-independent methods it is found that Coulomb and mass difference corrections are unable to cure this discrep­ancy.Optical potential models for antiprotonic atoms (J. Thaler)The strong interaction effect in atomic bound states of various antiprotonic atoms is cal- culated_from a basic two-particle interaction of the NN system and compared with available experimental data. A comparison is made with62results from the usual optical potential mod­el and the agreement with experiment is im­proved without any free parameters. The basic pN interaction is described by meson-theoret­ical elastic and phenomenological inelastic contributions, and the dependence of the re­sults on the annihilation model is discussed.Pion-nudeus physicsPion absorption by 3He: Two-body absorption contributions (C. Y. Cheung and O. V. Maxwell)Recently, the reactions 3He(ir“,pn)n and 3He(ir+ ,pp)p have received attention in thestudy of the isospin dependence of nuclear pion absorption. With the acquisition of new data at a pion lab energy of 65 MeV at TRIUMF [Moinester et al., Phys. Rev. Lett. 52, 1203 (1984)] to complement data obtained previous­ly at SIN and LAMPF, the behaviour of both the ir+ and i t -  absorption cross sections is now known empirically from near threshold up to the A-excitation region. In these experi­ments an outgoing proton and one of the other two outgoing nucleons are detected. By ad­justing the angle between the detectors ap­propriately and taking measurements at theright nucleon momenta, the two-body contribu­tion to the absorption cross section, where the third outgoing nucleon has small momen­tum and acts essentially as a spectator, can be extracted. Because the i t -  absorption must involve an initial isotriplet (to produce the outgoing pn pair), whereas both 1=0 and 1=1 absorption of a tt+ are possible, the ratio of two-body contributions to ir+ and ir~ absorp­tion is of fundamental interest.To date, theoretical studies of this ratio have concentrated on rescattering contribu­tions with an intermediate A33 resonance (diagrams B and C in Fig. 63) and yield values which are much too large [Lee and Ohta, Phys. Rev. Lett. 49_, 1079 (1982); Toki and Sarafian, Phys. Lett. 119B, 285 (1982)]. While the A-rescattering contributions domi­nate in the 1=0 channel at intermediate ener­gies, their contribution in the 1=1 channel is expected to be less important relative to other contributions. This suggests that cal­culations based on the A-rescattering dia­grams alone underestimate the i t -  absorption cross section relative to the ir+ cross section. To test this hypothesis we have initiated a study of pion absorption in 3He incorporating rescattering processes with an intermediate nucleon (diagram A, Fig. 63) and3He( A )(D )Fig. 63. Two-body contributions to pion absorption by 3He.a phenomenological s-wave rescattering term (diagram D), in addition to the A-rescat­tering contributions. Both the nucleon and A propagators are treated in a static, local approximation and the meson exchange propaga­tors are evaluated at a fixed energy transfer evaluated with all three initial nucleons at rest. A phenomenological monopole form factor is included at each internal meson-baryon vertex with a cut-off mass consistent with ird absorption studies or OBE models for the NN interaction. Preliminary numerical results indicate that rescattering diagrams with intermediate nucleon states are indeed rela­tively more important for ir~ absorption than for it+ absorption and that their inclusion substantially reduces the ratio of u+ to n~ two-body absorption in 3He.n-nucteus scattering (B.K. Jennings; N. de Takacsy McGill)The role of multiple scattering in ir-nucleus scattering has been a topic of much interest for many years. The problem is frequently discussed in terms of the LLEE effect for elastic scattering. The LLEE effect also plays a role in inelastic scattering, and particular inelastic scattering transitions are particularly sensitive to the multiple scattering. The inelastic 0+ to 0+ transi­tions at forward angles are zero at forward angles in Born approximation (i.e. go like the momentum transfer squared) while in DWBA the cross section tends to be forward peaked. The distribution of m substates in 0+ to 2+ transitions is also strongly dependent on the amount of multiple scattering. The currently available data tend to indicate that the mul­tiple scattering is small around T^ = 70 MeV and increases for both higher and lower ener­gies. The reason for this is currently63unknowns The work has helped motivate sever­al experimental proposals.Effects o f A propagation in nuclei (M.J. Iqbal)One of the most important pieces of informa­tion we have learned from studying ( p , i t ) ,  (p,Y) and other nuclear reactions at inter­mediate energies is the strong excitation of the A(1232) resonance in nuclei. Most of the calculations involving A excitation use a static approximation for A propagation, which is equivalent to using a closure approxima­tion. However, this approximation may only be justified near the resonance and may cru­cially depend on A width. By calculating A wave functions in a nucleus, using both a static and nonstatic A propagator, one can study the validity of static approximation. Present calculations suggest that the static approximation can at best be reliable near resonance. Detailed results will be pub­lished soon.The quark model and QCDQuark structure effects in the a-a transition (J. Thaler)a-a phase shifts for L=6 are experimentally well defined and mainly sensitive to the 2tt— exchange interaction. Further they are most­ly sensitive to the a-a interaction at a separation of about 4-5 fm, where the two nuclei have little overlap and many-particle effects are negligible. Realistic NN poten­tials like the Paris potential are not able to describe these data. We have found that the description of the nucleons as bound states of quarks leads automatically to a considerable enhancement of the 2ir-exchange interaction between two a-particles. The resulting a-a interaction describes the ex­perimental data with a surprising accuracy.Vector meson decays in chiral bag models (O. V. Maxwell and B.K. Jennings)Vector meson decays have been examined in a model where a confined quark and antiquark annihilate, producing a pair of elementary pseudoscalar mesons. Two versions of the pseudoscalar meson-quark interaction were em­ployed, one where the coupling is restricted to the bag surface and one where it extends throughout the bag volume. Energy conserva­tion was ensured in the model through inser­tion of an exponential factor containing thebag energy at each interaction vertex. To guarantee momentum conservation a wave packet description was utilized in which the decay widths were normalized by a factor involving the overlap of the initial bag state with the confined qq” states of zero momentum. With either interaction the model yields a value for the o meson width that is a factor of two larger than the empirical one. For the K and <t> mesons the computed widths depend strongly on the interaction employed.Meson-nucleon scattering (B.K. Jennings; E.A. Veit, Porto Alegre, Brasil; A.W. Thomas, Adelaide; R.C. Barrett, Surrey)This is part of a four-continent collabora­tion studying the scattering of baryons and the pseudoscalar octet of mesons. The moti­vation of this work was to check the cloudy bag model (CBM) and to see if it gives any new insights into the scattering processes. One of the first conclusions of this work was that it is necessary to use the volume coup­ling version of CBM rather than the surface coupling version in order to describe the s-wave scattering. The volume coupling ver­sion of the CBM has a contact term (second order in the meson field) which has effects very similar to the vector meson exchange of meson exchange theories and gives the correct tt-N scattering lengths in Born approximation. The CBM works quite well for KN scattering although the spin-orbit force is too weak. The strengthening of the spin-orbit force ap­pears to arise from many small effects - re­normalizations, recoil, centre-of-mass cor­rections, etc.The model works less well for ttN than KN. In the tt-N scattering, particularly s-wave, there is clear evidence of long-range attrac­tion and short-range repulsion, both missing in our current version of the CBM.Meson physics in the quark model (S. Godfrey; N. Isgur, Toronto)Quantum chromodynamics (QCD) is considered to be the correct theory of the strong interac­tion. However, it has proven to be extremely difficult to extract predictions from the theory which can be compared to experiment. At the same time we have a very successful phenomenological model of hadron physics, the quark model, which in its recent manifesta­tions has been made more sophisticated by in­cluding the dominant features expected from QCD: a one-gluon-exchange potential at short distance and a long-range linear confining64potential. Such models are very successful in explaining the spectroscopy of heavy quarkonium systems (the and T systems).In order to extend these models so that they can treat all hadrons, including those made of light quarks, in a unified framework we have included relativistic effects and the effect of the QCD running coupling constant. The resulting model [Godfrey and Isgur, Uni­versity of Toronto preprint (1984)] gives reasonably good agreement with all known meson masses and so provides strong evidence that all mesons from the lightest isovectors to the heaviest beautyonium states are gov­erned by the same physics (see Fig. 64). We have also completed an extensive decay analy­sis including strong decays, electromagnetic transitions and some weak decays. The decay analysis gives further support to the basic quark model assignments.One current project is to extend this analy­sis to include isospin-violating effects in mesons. Another is to study high-spin mesons which will help to understand the nature of confinement at large distance.I3Si  T  V ---- £  K* pI'Sq---- \----------- Vc------------7 7 ,77' KFig. 64. A graphic illustration of the uni­versality of meson dynamics from the it to the T, showing the splittings of 3P2 and -^ Sq from 3Sj in the bF, cc”, si”, us and ucT families.consistent with those predicted for the L=3 ss meson with jPc=2++ [Godfrey et al., Phys. Lett. 141B, 439 (1984).This observation demonstrates the importance of understanding the more prosaic 'old1 phys­ics of QCD before looking for new physics interpretations of unexpected phenomena.Semileptonic decays o f mesons (S. Godfrey)An important puzzle in high-energy physics is the relation between the quark eigenstates of the weak interaction and those of the strong interaction. This relation is expressed as a mixing matrix, the most popular parametriza- tion being that of Kobayashi and Maskawa. To understand this matrix and eventually the nature of the mixing it is important to know the values of the various entries. These can be found by comparing predictions for the weak decay amplitudes and widths with the measured values. However, for these compari­sons to be meaningful we must have reasonably good estimates of the hadronic wave function effects and of the phase-space integrals.To evaluate the meson semileptonic decay amp­litudes I have used the meson wave functions of a very successful model of hadron physics [Godfrey and Isgur, University of Toronto preprint (1984)]. The calculation includes relativistic effects and gives reasonably good agreement for the form factors[Godfrey, University of Toronto preprint (1984)]. It is hoped that the results will be useful in extracting the Kobayashi-Maskawa mixing angles from measured data. The calcu­lations are to be extended to nonleptonic weak decays and to CP violation in the K°-K° system. If those results are reasonable the analysis will be further extended to relate bounds on rare decay processes to bounds on new physics.Coulombic confinement model o f quark dynamics (VJ. Wilcox; O. V. Maxwell, Florida International University)The £,(2.22): An L = 3  ss meson? (S. Godfrey; R. Kokoski, N. Isgur, Toronto)The recent discovery of the £ resonance in the decay y£ by the MARK III group atSLAC has fueled considerable speculation on whether it could be the long-sought-after Higgs particle of the electroweak theory. We point out that the mass, decay widths and production cross section of the £(2.22) areThe authors are continuing to investigate the phenomenological consequences of the Coulomb­ic confinement model of quark dynamics. This model is applicable to heavy-light quark sys­tems (or heavy-light-light baryons) and simp­ly views such states as relativistic, con­fined hydrogen-like (or helium-like in the case of baryons) quark systems. We have ap­plied the model to charmed mesons and have calculated the spectra and transition rates65for D* ■+ Dy , D* -*■ Dir and F* Fy . Several striking results have been obtained. First, we find a description of such states which agrees with the best theoretical estimate for quark self-energy and Casimir energy effects. Second, this description naturally yields a reduced effective coupling constant (ag ~0.4), restoring our confidence in the appli­cation of perturbation theory in bags. Third, the model makes a theoretical prediction for the transition ratio (D*° -»■ D°y )/D*+ ■* D+y ) which is substantially lower than values ob­tained from more traditional bag models where both quarks are allowed to move through the entire bag volume. We are at present extend­ing the model to the case of baryons, where one must calculate the inter- (dynamical) quark Coulombic and magnetic interactions.Lattice quantum chromodynamics (W. Wilcox, R. Woloshyn)It has become increasingly clear during the past several years that numerical study of gauge field theory on a lattice (a finite, discrete space) can provide valuable insight into quantum chromodynamics (QCD), the theory of strong interactions. Last year a program for lattice QCD was initiated at TRIUMF. The focus of this program has been the study of lattice hadron structure. The goal is to learn more about gauge field dynamics and, ultimately, to provide empirical tests of QCDThe first results of this program have been submitted for publication [TRI-PP-84-90 and 84-91] .It is widely believed that lattice QCD can describe the properties of physical hadrons. It is important, therefore, to study the structure of lattice hadrons and see if it agrees with our theoretical and empirical ex­pectations. We have carried out a detailed study of a direct correlation technique for probing lattice hadron structure which in­volves various gauge field (gluon) and fermi- on (quark) operators. We have looked at mesons in a model with SU(2) colour to save computer time. With gluonic probes we can discern changes in chromomagnetic fields and flux in the vicinity of a meson. This is similar to what has been observed for lattice glueballs. Fermionic probes look promising but, because of limited computing power, we were not able to gather enough statistics to make a definite statement about the size of lattice mesons using this technique.An indirect probe of hadron structure is pro­vided by placing the hadron in an external electromagnetic field and observing the in­duced dipole distortion (polarizability). We have done this for electrically neutral mesons (again with SU(2) colour) by calculat­ing the mass of the meson in the presence of an external electric field. At low external field the expected decrease in mass, associ­ated with dipole polarization, is observed (Fig. 65). Qualitatively the results agree with what is expected in the nonrelativistic quark model.Our present calculations are concentrating on electromagnetic properties and transitions of hadrons. These have always been an important testing ground for theories of strong inter­actions but only very little of this has been done on the lattice. The construction of the electromagnetic current on the lattice is relatively straightforward and recently tech­niques which make the calculation of vertex or three-point functions feasible have been developed. The calculation of the pion elec­tric form factor has started and appears to be feasible with our limited computer resources. Radiative Ml decay rates (e.g. o tty), calculable using the same techniques as in the form factor calculation, are being considered.Fig. 65. Pseudoscalar mass (in lattice units) versus the electric field parameter 77. The solid line is the expected behaviour of the mass assuming linear response and and a value of the electric polarizability equal to the central value of the experimental limit on the electric polarizability of ir~. The dashed lines use the extreme values of the experimental limit.66High-energy physics and weak interactionsToponium-Z° mixing and anomalous events at pp collinders (J. Ng; L.N. Chang, VPI)Recently the UA1 experiment at the CERN pp" collider has discovered events that are con­sistent with the existence of the t quark with a mass between 30 and 50 GeV/c2. This gives a mass range for the toponium of 60 to 100 GeV/c2, thus overlapping with the Z° mass. In general when two levels with the same quantum numbers are close to each other they mix. We have to investigate this mixing between the s-wave and p-wave toponium states and the Z°. It is found that the p wavehas a larger effect on the Z°. The massshift of the Z° due to this mixing is very small. We found that it contributes only 0.5% to the p parameter. On the other hand the mixing induces a large effective toponium-Z°- photon vertex. In the case of near degenera­cy, i.e. the p-wave toponium T^ and Z° mass have a mass difference less than that of the binding energy between the two quarks, the ratio of the production T^ x the branching ratio of T^ ■+• ISLy to the production of Z° x the branching ratio of Z° + i l  is 0.16. This is the upper bound for this ratio and it drops rapidly as the mass separation between T^ and Z° increases. This in turn leads us to expect drastic changes in our expecta­tion of the characteristics of the toponium. We are now investigating this aspect.Higgs boson production in polarized e'e annihilation (J. Ng; R. Bates, UBC)It is anticipated that if the Higgs boson has mass in the range between 2Mt and 2MZ, where Mt and Mz are the masses of the t quark and the Z° boson, respectively, the cleanest way to find it is in e+e- colliders such as LEP and/or_SLC. The reaction to look for is e+e~ where I  denotes e, u or v andthe H° will subsequently decay into two t quarks which fragment into two jets. How­ever, one of the backgrounds comes from e+e“ ->• Z ° l i  which has a larger rate and the Z° has a substantial branching ratio into two t quarks. Using the polarization of the ini­tial e+ and e~ beams we find that one can substantially cut down the background due to gauge boson production such as the one above. Details of the characteristics of the H°, such as energy spectrum and angular distribu­tions, are now being calculated.A rare decay o f the kaon (J■ Ng)Recently experiments are being mounted to measure rare decay modes of the kaon to un­precedented levels. Examples are Kql ue and K+ ■+ ir+vv. In the latter reaction an impor­tant process which can also be studied is K+ + tt+yy which currently has only an upper limit. We are calculating this last reaction in the standard model and investigating its sensitivity to the t-quark mass and the other parameters of the minimal six-quark model. We also extend the study to probe effects of heavy fermions and other effects arising from generalizing the standard Weinberg-Salam model. Since the branching is of the order 10~6 we expect this reaction to be a good handle on which to study the effects of pos­sible technifermions and effective flavour- changing neutral currents not contained in the standard model.Spin-0 boson production Z° decays and tests for compositeness o f Higgs bosons (J. Ng, P. Zakarauskas; P. Kaiyniak, Carleton)Recently there are nonperturbative lattice calculations on the parts of the standard model involving the scalar potential that indicate the Higgs boson must be light com­pared to 1 TeV in order that the theory be consistent. For a light enough Higgs boson the cleanest way of producing it will be at LEP using Z° H°£.+f.“ decays. Experiments done by OPAL will be crucial in settling the question of whether Higgs bosons exist below the mass of the Z°. There exist other schemes of mass generation such as techni- colour scenarios that replace the Higgs boson in gauge theories. Typically they predict both parity even and parity odd bosons. Also, composite models of gauge bosons have spin-0 resonances, with both parities. Recently we have addressed ourselves to the question of distinguishing between these possibilities in Z° decays. We calculated the scattered plots of the energies and invariant mass of the dilepton pair and showed that Higgs pro­duction can be separated from technihadrons but it is harder to distinguish it from com­posite spin-0 hadrons. The distinction between technipion and composite pseudoscalar is even smaller. This is of great importance to experimentalists as well as to the study of the models.To further separate the cases one needs to calculate the Q2 behaviour of the form67factors involved. Only the standard model is which is not a gauge theory, also exhibitswell defined enough to do that. We are pro- spontaneous supersymmetry breaking. In orderceeding with the calculation of the Higgs- to illuminate the physics of this very impor-photon-Z° and Higgs-Z°-Z° couplings to the tant phenomenon of dynamical supersymmetryone-loop order. This knowledge is essential breaking, it is important to calculate morefor detecting the compositeness of scalar rigorously the Witten index in this model,particles even when they are not a fundamen­tal field in the theory.Muon spin rotationCharged Higgs boson effects in muon decays, ve-eiectronscattering and e ’ e annihilation Skewed field technique in muon spin rotation(J. Ng) (R.E. Turner)We analysed the effect of charged Higgs bosons in e+e“ y+ii~ and muon decays. Con­straints in the parameter space of the mass of the boson versus the strength of the Yukawa couplings are obtained. Current pre­cision measurement of muon decays rules out charged Higgs of mass less than 100 GeV and couples with strength of the order of e. Novel interference effects between charged Higgs exchange and standard gauge boson ex­changes are found in '■v^ e scattering.Field theoryPossible supersymmetry breaking in 1 +  1 dimensional supersymmetric quantum electrodynamics (J. Ng; L.N. Chang, VPI)The massless Schwinger model, i.e. QED in 1+1 dimensions, is known to break supersymmetry via the anomaly and the photon gets a mass dynamically. It is of interest to study whether this so-called Schwinger mechanism remains after supersymmetry is introduced into the system. We have constructed the Lagrangian for this case using the superspace technique; as in the case of four-dimensional supersymmetry the fermion has a scalar part­ner and the photon obtains a spin-1/2 Majorana partner as well as a scalar partner. We have constructed the Lagrangian in terms of these physical fields using the Wess- Zumino gauge. We also notice that by count­ing the number of bosons and number of ferm- ions in the ground state the Witten index Tr(-l)^ vanishes, signalling supersymmetry is broken in the theory.We are calculating the quantum fluctuations in this theory to obtain a better under­standing of this phenomenom. It is expected that this symmetry breaking should reveal itself in perturbation theory as indicated by Witten. This is the first gauge theory where supersymmetry is dynamically broken. Pre­vious study using the Wess-Zumino model,For an arbitrary external field direction three distinct muon-spin relaxation functions can be defined [Turner, Phys. Rev. B (in press), TRI-PP-84-12]; that is, there is a longitudinal relaxation function in the di­rection of the external field and two trans­verse relaxation functions in the plane per­pendicular to the field direction. One of these latter functions, termed the coplanar transverse relaxation function, lies in the plane defined by the field direction and the incoming muon spin polarization, while the other, termed the perpendicular transverse relaxation function, is perpendicular to both the field direction and the incoming muon spin polarization. All three relaxation functions can be measured simultaneously if the applied external field is not in either of the standard geometries; that is, not parallel to the incoming spin polarization (longitudinal geometry) nor perpendicular to it (transverse geometry). This suggests that a skewed field arrangement provides an exper­imental technique in which the traditional relaxation functions for a given sample may be determined in a single experiment using a single apparatus. Thus, at the very least such an experimental alignment eliminates the downtime required when changing from one geometry to the other in the determination of relaxation functions for a given sample.To illustrate this experimental arrangement [Turner, TRIUMF preprint TRI-PP-84-86] the dynamics of the spin polarization of a muon which has thermalized in a solid is expressed in terms of the crystal frame spherical har­monic moments of the local classical random magnetic field distribution associated with static Kubo-Toyabe theory. This local field distribution of, in general, arbitrary symme­try is a classical approximation to the mag­netic field interaction between the spin of the muon and the spins of the nuclei associ­ated with the site at which the muon has stopped. The moments of this classical dis­tribution are equated with the quantal68moments generated by this magnetic field while the resulting spin polarization of the muon is related to the observable longitudi­nal, coplanar transverse and perpendicular transverse relaxation functions. If the sam­ple is a powder only the monopole expansion coefficient of the classical field, which exists for all general sites, contributes to the observable relaxation functions whereas for single crystals the dynamics contains all the moments of the classical field. In the latter case the existence of dipole and quad- rupole symmetries (expansion coefficients)can be elucidated using the zero field tech­nique while higher-order moments require an applied external field. For example, tetra­hedral and octahedral sites in body-centred cubic crystals can, in principle, be dis­tinguished in zero field experiments because their isotropic and anisotropic quadrupole moments are of the same order of magnitiude. The explicit nature of the local classical magnetic field, whether it is, say, a Gaussian or a Lorentzian, is not required in this elucidation.69APPLIED PROGRAMS DIVISIONINTRODUCTIONThe applied program evidences maturity, mid age and youth in its programs. The radio­therapy program continues its clinical trial phase and now is planning a channel upgrade to improve the reliability and flux on its 10-year-old pion channel. Sixty-nine patients have been treated. Now that the CP-42 cyclo­tron is operating on a full beam schedule even while expansion takes place, both AECL and PET have received regularly scheduled radioisotope delivery. AECL has produced an ultra-high purity 123I using the CP-42 cyclo­tron. PET has produced 18F-labelled 6-fluoro- dopa that is qualified as a PET scanning agent and it is in regular use. Animal stud­ies by the TRIM group demonstrate excellent heart-imaging quality for 123i lactone de­rived fatty acids and patent applications have been made for this work. Beam line 2C work is still proceeding and TRIM research with new radioisotopes waits for the beam line commissioning.BIOMEDICAL PROGRAMThe activities for the pion radiotherapy program in 1984 are very similar to those of1983. The average cyclotron current output was about 10% higher than the previous year, and a total of 112 days were available for treatment. This was divided into 3 blocks of 6, 12 and 6 weeks duration in January-February, May-August and October-December, respectively. The total number of weeks available for treatment is roughly the same as last year, but the two short runs of 6 weeks were plagued with extensive breakdowns (of up to five days in duration), seriously reducing the efficiency of patient treatment and accrual. Hence, only 6 patients were treated in each of the 6-week blocks compared to 18 patients treated in the 12-week block. A total of 15 brain patients were treated with doses varying from 30 Gy to 33 Gy in 15 fractions. Most of these patients were treated with pions only (instead of pion boost on photon as in previous years), except during the last run all the brain patients received 2 or 3 dose fractions in photons because of the breakdowns in the cyclotron and beam lines. The dose-fractionation sche­dules for pelvic patients were very similar to those of last year, but larger and larger tumours were being treated. In the last runa record size of a 13 cm diameter tumour was treated, which required 1.5 h of irradiation time even for the most stable beam. Thirty patients were handled in the past year (Table XI), making the total number of patients treated since May 1982 to be 69.It was known from the beginning that patients treated with pions are quite radioactive (1-3 mrem/h on contact right after treatment). This radioactivity has a half-life of about 10 to 20 min and it is believed to be coming mostly from the 1J-C produced by pions stop­ping in tissue. In order to reduce the ex­posure to the biomedical personnel and family members the patients were isolated in a shielded room for about half an hour after treatment. However, many of them were radio­active enough to activate the TRIUMF vehicle gate radiation alarm on exit. During the last year this radioactivity was put to good use in a collaboration with the TRIUMF-UBC PETTable XI. Summary of TRIUMF pion patient treatment for 1984.RunFraction Patient finished/intendedTotaldaysDose/fraction (n- rad)Jan.-Feb. X brain 15/15 18 170X bladder 10/10 11 250X skin 10/10 11 250X prostate 10/10 15 250X bladder 10/10 17 250X rectum 10/10 10 250May-Aug. X brain 15/15 21 200X brain 15/15 21 200X brain 15/15 21 220X bladder 15/15 21 250X brain 15/15 20 170X prostate 15/15 19 250X brain 15/5 22 170X bladder 12/12 15 250X bladder 10/10 8 250X brain 15/15 20 220X rectum 12/12 14 250X brain 15/15 21 200X rectum 10/10 15 250X brain 15/15 17 220X brain 15/15 20 220X rectum 10/10 13 250X brain 15/15 20 220X prostate 15/15 27 250Oct.-Nov. X brain 13/15 23 200X brain 13/15 22 200X prostate 12/12 13 250X brain 15/15 26 200X brain 13/13 17 200X rectum 10/10 10 25070team to visualize the pion stopping region by mapping the positrons emitted from 1JC pro­duced on pion capture. Several brain patients were taken to the Acute Care Hospital on the UBC campus for PET scans immediately after pion treatments. The trip usually took 20 min but there was still enough positron emission for a good scan. From these pion-induced ac­tivities the actual spatial dose distribution in patients can be reconstructed and will provide important input to treatment planning based on CT information.Minor hardware breakdowns on the M8 channel appear to be on the rise for the last year. The problems include vacuum leaks, water leaks, power supply overheating and electri­cal short-circuiting. A total of 2 treatment days were lost for the entire year. This amount of breakdown is not unexpected for a 10-year-old channel but the M8 channel is particularly vulnerable. Because of its geometry the channel has to rest on movable blocks instead of the concrete floor like the rest of the TRIUMF secondary channels, and hence is subject to vibrations and other dis­turbances whenever there is activity around the 1AT2 area. Furthermore, because of the high neutron background in our treatment room we were encouraged to fill all the gaps in the channel with small concrete blocks which are quite incompatible to good maintenance on the wiring and plumbing of the channel. About a quarter of these blocks were removed during the last year.We still experienced a high level of radioac­tive gas build-up in our treatment room in the last year, although the personnel dosecontributed from these gases is relatively insignificant compared to that from the neu­trons . An air supply duct was installed in the treatment room, which is now being sup­plied with 120 ft3/min of conditioned air.This has greatly improved patient comfort during treatments (the treatment room was either too hot or too cold), and the slight pressure build-up in the treatment room also helps to counter the flow of radioactive gas into the control room during patient treat­ment .Various efforts were made to upgrade the pion flux in the M8 channel. A pion productiontarget test was conducted in January 1983 at1AT1 using a thin diamond target, and a 50% improvement per unit length was obtained with the synthetic diamond as compared with graph­ite. However, a similar target test in June using a thick diamond target at 1AT2 yieldedonly a 4% increase over the standard berylli­um target. It appears that the advantage of the higher density of a diamond target was nullified by the higher rate of beam degrada­tion in the thick target, together with a relatively large decrease of cross section at lower proton energy. Tests for permanent mag­netic materials suitable for the M8 channel were also conducted by the Experimental Facilities group. After several rounds of discussions a proposal was presented to TRIUMF management for the installation of a prototype permanent quadrupole next year.Some rn vitro biological experiments were made on the channel during the evening after patient treatment. From June to December a visiting group of physicists from the Uni­versity of Surrey conducted a series of pion capture ratio experiments using various bone equivalent materials. A system of scintil­lation counters and a GeLi detector on loan from the University of Victoria were set up on a mobile stand in the patient treatment room. These measurements are an extension of an experiment that was conducted on the same pion channel on tissue equivalent material over four years ago.42 MeV CYCLOTRONStatusThe end of 1984 marked the completion of the first full year of cyclotron operation. The machine was used approximately 90% of thetime for radioisotope production for AECL Radio-Chemical Co. The remaining 10% was used for the production of positron emitters for the PET program at the UBC hospital.To make it possible to use the target cavefor its intended purpose - to accommodate a switching magnet and a number of targets - the building was expanded by a new service room and enlarged control room. The building extension was virtually completed by the end of the year.OperationThe cyclotron operated every week of the year and delivered a total of 454 mAh of beam. Figure 66 illustrates the weekly beam produc­tion and shows a gradual increase in produc­tivity over the year. The low productivity in week #6 is not real, because that was aweek of only 3 days due to a switchover froma Thursday-to-Wednesday week to a regular712 0  2 5  3 0  3 5  4 0  4 5  5 0Week15000 —St 9 0 0 06 0 0 0  -52 0  2 5  3 0Week1 /zA Hour=3.6X10"* C o u lo m bFig. 66. Weekly 42 MeV beam delivery.calendar week. The low productivity in and just before week #39 was real and was caused by a combination of circumstances: higherthan usual component failure rate, instal­lation of a new 27 MeV extractor arm and the ever-increasing impact of the building opera­tions .During the 2nd and 3rd quarter production was also limited by the AECB licence, which did not allow beam currents higher than 150 pA and up to a charge limit of 10 mAh/week on the solids targets. Subsequently the beam current limit was increased to the requested 200 pA and the charge limit, which had caused the cyclotron to be shut down for about 30 h, was removed.The cyclotron was usually run for four 24 h days plus a few hours or approximately 100 h/week, from Friday to Tuesday morning. Subsequently there was a 20 h cool-down peri­od, one day for maintenance and improvements and one day to test the machine for the next operating period and, initially, for operator training. Deviations from this basic sche­dule were necessary from time to time due to building construction, licence limitationsand/or equipment failures during the early part of the week that had resulted in can­celled runs. With an operating and mainte­nance crew of eight it has been possible to run all shifts outside normal office hours with one person on duty and with from four to six available during the maintenance days.Table XII lists the scheduled and actual beam delivery for each user. The 'efficiency of beam delivery' compares the actual with the scheduled quantities. Ideally this perform­ance indicator should approach 100%. In prac­tice this will not happen very soon, as we always schedule more beam than can reasonably be expected within the available beam current and time limitations. The readily available beam current on target is 150 pA. When the machine runs well 200 pA can be achieved.Table XIII lists the scheduled, cancelled and actual runs for the year. Cyclotron beam schedules are drawn up for each week, in consultation with the users, one week before the start of the schedule. The construction activities caused a substantial number of cancelled runs, mostly due to high radiation fields when the soil around the building had been excavated to accommodate the expansion. User-cancelled runs were rather frequent in the beginning of the year because of the developmental character of their operations. User-added runs are accommodated whenever possible, and especially after a run can­celled due to machine failure.Table XIV shows how cyclotron time was dis­tributed among the users, along with downtime due to failures and time for start-up and tuning. Although the intent has been to run the cyclotron 100 h/week, the average real­ised for 1984 has been 84 h/week. However, it increased from 64 h/week during the first quarter to 98 h/week during the last quarter. The gradual increase was due to the comple­tion of recruiting and training of the eight- member crew by July 1 and the substantial completion of the more disruptive construc­tion operations by September.Building extensionThe building for the CP-42 cyclotron was originally designed primarily for commercial radioisotope production at two fixed energy beam ports and at the end of the variable energy beam line. The space required for the necessary equipment was designed in consulta­tion with the Cyclotron Corporation, the manufacturer. Provision for the production72Table XII. 42 MeV beam productionAECL PET TotalScheduled beam delivery (mAh) 555.3 2.2 555.5Actual beam delivery (mAh) 452.1 1.4 453.5Efficiency (%) 82 64 82Table XIII. Breakdown of cyclotron irradations (runs)AECLIsotopeproductionIsotoperesearchPETPatientresearch Other Totala Scheduled 359 137 239 8 743b Added by user 75 66 27 4 172c Cancelled by user 22 34 53 3 112d Cancelled due tocyclotron failure 33 34 23 1 91e Cancelled due to gastarget failure 1 14 4 19f Cancelled due to solidstarget failure 5 0 2 7g Cancelled due toconstruction 10 0 10 1 21h Actual runs 361 119 170 7 657Rate of cyclotron successh/(h+d) % 92 78 88 88 88Table XIV. 42 MeV cyclotron time breakdownAECL PET Totala Start-up and tuning (h) 221 127 348b Beam on target (h) 3336 100 3436c Downtime (h) 529d Unaccounted for 63Scheduled cyclotron time a+b+c+d 4114 262 4376Running efficiency b/(a+b) (%) 94 44 91Cyclotron availability a+b/(a+b+c) (%) 8873I   1Fig. 67. The 42 MeV proton irradiation facility, before (top) and after (bottom) the building expansion. 1) Cyclotron vault, 1A) cyclotron, IB) 27 MeV solids target station, 1C) 26 MeV gas target, 2) variable energy target cave, 3) service room, 4) active waste storage, 5) active waste holding tank, 6) cooling equipment, 7,10) personnel change and monitoring areas, 8) power supply room, 9) control room, 11) rabbit tubes to radiochem­istry hot cells. At year-end the building was complete, but the new beam switchyardstill has to be installed in the target cave.of positron emitters for the PET program was made at the end of the variable energy beam line. This arrangement is in use at present and is shown in Fig. 67, (top).The PET gas target can only receive beam when there is no target in the solids target sta­tion, and when the solids target station receives the beam the PET target cannot be used. The switch-over procedure takes a fair amount of time, so the need for a switching magnet soon became apparent. Because the switching magnet could accommodate up to 9 beam lines, it would also allow rapid switch­ing to any one of a series of PET and AECL targets.The building had now become too small for a variety of reasons:1) The Cyclotron Corporation supplied more voluminous power supply and controls equip­ment than originally specified, causing over­crowding in the power supply room.2) The TSG-designed area access control equipment took up more space than expected, adding to the congestion of the control room and aggravating overcrowding in the power supply room.3) The maze giving access to the target cave proved inadequate and required an extension which rendered half the target cave areauseless. This was aggravated by the rela­tively high radiation field in the target cave, because it was impractical to shield the solids target station, which required frequent repairs.4) Space for a switching magnet and its power supply was not allowed for in the original building design.5) No space was allowed for a properly equipped change area.Plans for a building expansion to be con­structed as part of the meson hall east ex­tension were drawn up and approved early in1984. Construction started on July 1 and the extension was essentially complete by Decem­ber 31, about two months behind schedule.Figure 67 (bottom) represents the building as completed. A new service room was provided that will accommodate some of the equipment presently in the power supply room, new equipment such as the switching magnet power supply, a new change area for the target cave, a future target handling cell and grain analysis counting equipment. The old maze has been abandoned and a shielding door provides access to the target cave via the service room. This has considerably improved the cave shielding and reduced the radiation field in the control room. The door has also made the maze extension obsolete thus freeing the whole cave for switching magnet and targets. The power supply room will now have more space for a properly sized change area once the area access control equipment and the solids target system controls have been moved to the new service room. The control room has been doubled in size and now provides space for switching magnet and beam line controls.The building construction had a major impact on cyclotron operation. As a comparison be­tween Fig. 67 top and bottom illustrates, several areas of concrete wall, some as large as 12x12 ft, had to be removed. Excavation of the soil for the new area, and the subsequent removal of wall, restricted operation of the targets in the target cave to week-ends and days when the contractor had to be barred from the construction site. The danger of overexposure of cyclotron and construction crews was a continuous concern as were the normal hazards associated with new construc- *A report 'Conceptual Design for a TRIUMF Grain Analysis System' was produced during the summer by J. Elder, UBC Engineering co-op student, which discusses the possibility of protein analysis using the 42 MeV cyclotron.tion: dust, noise, water, all detrimental to the operation of a cyclotron.Yet, thanks to careful scheduling and due to the excellent co-operation between contractor TRIUMF's civil engineering group, plant group safety group and 42 MeV operations crew, it has been possible to continue operation of the cyclotron with only minimal disruption and continually improving productivity.ImprovementsThis first year of cyclotron operation has provided the necessary experience to assess the reliability of the cyclotron and target systems. Particularly during the first half of 1984 the failure rate has been quite high. This is not too surprising, as rather than a proven commercial product this cyclotron actually was the prototype for the Cyclotron Corporation's line of CP-42 cyclotrons. Although one cyclotron was commissioned ear­lier than ours, most of the heavy duty test­ing was done on our machine. As a result of this testing and resulting makeshift improve­ments, several components should actually have been changed before installation at TRIUMF. This was never done, presumably because the company's financial situation was deteriorating, resulting in receivership early in 1983 and finally complete dissolu­tion of the company in January of 1984.As a result the list of improvements, some still in the planning stage and some started in 1984, is too long to include in this re­port. The most urgent improvement, new sol­ids target stations (see Fig. 68) has been completed to the testing stage. As the cyc­lotron has to be kept running to maintain the radioisotope production program, it will take some time before all improvements can be implemented.Facility expansionThis year the cyclotron served two solids target stations and two gas targets. The solids target station and gas target in the cyclotron vault are served by separate fixed- energy beam lines and used for commercial radioisotope production. The variable energy beam line serves both a solids target station for commercial radioisotope production and a gas target for the production of positron emitters for the PET project. The latter two are located in the target cave and in series in the beam line. The solids target has to be removed from the station to allow beam to reach the gas target.Fig. 68. Remote-controlled target station. 1) target chamber, 2) target port, 3) beam collimator, 4) target, 5) rabbit for target transport, 6) air tube swing terminal, 7) air tube to hot cell, 8) target cooling.With the building extension now completed we will be in a position to expand the target facilities in the cave by installing a switching magnet which may eventually serve up to nine targets.RADIOISOTOPE PROCESSING (AECL)The establishment of a 500 MeV beam schedule having lengthy periods without high intensity beam has continued to put pressure on the ab­ility of AECL to provide the service that the medical community requires for radioactive isotopes used in routine diagnosis and re­sulted in a loss of potential 82Sr business. However, because of a back-up arrangement with Brookhaven National Laboratory purchases of 127Xe ensured that contractual commitments were met.On April 1 the Atomic Energy Control Board issued a production licence for the CP-42 cyclotron. By year-end shipments of 57Co, 67Ga, 111In, 123I and 201TJt were routine. The programme to improve the reliability and pro­ductivity of the CP-42 cyclotron put a lot of strain on the scheduling of all isotopes, but all customer requirements were met and byyear-end there was a significant increase in weekly isotope production. Three planned 123I production runs were cancelled because of cyclotron availability.J11In is being used for labelling monoclonal antibodies. For this application the product must be free of trace metal contamination which could compete for the labelling sites. Initial results indicate that the product produced at TRIUMF is already as good as any competitive material. Steps that would im­prove the product further have been identi­fied and will be employed should it prove necessary.The highlight of the year was the successful operation of the high power gas target for 123I production. Following improvements to the window cooling system in March, there were no problems associated with the gas tar­get. Improvements to the yield in September permitted the production of batches of 2-3 Ci of 123I. At the same time purity was im­proved and no other production method can yield such a pure material - >99.9% ^23l 24 h after production.500 MeV RADIOISOTOPE PRODUCTIONProton irradiation facility for solidsThe year 1984 was the fifth year of operation for the 500 MeV isotope production facility. The facility performed without failures and received 172 mAh, which is 21% more than in1983. The use of the facility is illustratedin the following:TargetsmAh to Targets deliveredYear facility irradiated to AECL1980 51 40 261981 56 53 381982 120 49 421983 142 70 741984 172 82 57The following is a breakdown of the kind ofradioisotopes produced in 1984:Number ofNumber of targetsTarget targets deliveredmaterial Isotope irradiated to AECLCsCSL 127Xe 73 49In l°3Cd 8 8As 58Ge 1 076Discrepancies between the number of targets irradiated and the number delivered to AECL are due to delays in delivery at the begin­ning and end of each calendar year and be­cause irradiated targets are counted twice if they include the year change.Gas targetThe 18F gas target failed in May soon after it was installed due to misalignment with respect to the beam. This target was an im­proved version of the first target, which failed due to leaks that developed around the windows. The welding design was changed and this time both the target's cylindrical hous­ing and the windows were made of inconel. The failed target was removed from the beam line in September and inspected and repaired in a hot cell. The repaired target has only one target chamber instead of two because the burst set of helium windows could not be re­paired. Subsequently the target was rein­stalled.Since October this target has been a valuable source of 18F for the PET program. For the time being it is used at a reduced pressure of 7 atm. The design pressure is 10 atm.Neutron activation analysisThe January shutdown was utilized to install a new Rotirad (rotating NAA capsule irradi­ator). Before installation the old Rotirad, which had developed a leak, was inspected in a hot cell and the nature of the failure was established. The redesigned device will not likely develop a leak of the same nature again. The original device, which had to be replaced when the Mkll neutron target was in­stalled, did not have the problems of its successor because it was made of stainless steel rather than of aluminum. Although it is more difficult to weld aluminum has a lower cross section for neutron capture.Towards the middle of the year the meson hall east construction necessitated relocation of the NAA rabbit receive stations because the Novatrack trailer complex had to be moved away from the main building. The receive station for the conventional rabbits and the Rotirad rabbit rotator are now located inside the building, in a small room specially built for this purpose behind the 500 MeV irradia­tion facility. In the process the neutron spectroscopy experimental room, now located partially under the new rabbit receiving room, was soundproofed and fitted with a door. Allthis has resulted in improved facilities for Novatrack and the neutron spectroscopy group.POSITRON EMISSION TOMOGRAPHYThe PET program continued in 1984 with sup­port from further instalments of Medical Re­search Council Special Project Grant SP-7. The positron emission tomograph built at TRIUMF in 1982 continued to work in exemplary fashion, a tribute to the maintenance of the machine by the PET and Electronics groups at TRIUMF.A considerable expansion occurred this year in the software package available for pro­cessing images from the machine, by means of the VAX 11/750 computer system in the Elec­trical Engineering Department at UBC (to which the tomograph is linked by a fibre op­tic cable). The complete coverage of the brain afforded by the seven axial images pro­duced by the UBC tomograph permits the con­struction from these data of coronal and sagittal images, providing views of the brain at right angles to each other and to the original axial images, although arbitrary choice of angle is possible. Coupled with correlation of PET function images with ana­tomical images from CR and MRI scanning, this represents a powerful battery of techniques for brain functional analysis. Further soft­ware permits the placing and manipulation of region-of-interest outlines on the images and the quantitation of brain biochemical func­tion within such regions in absolute terms.During the year the development of radiophar­maceutical scanning agents for PET continued in the TRIUMF radiochemistry annex. The pro­cess employed to produce 18F-labelled 2-fluo- ro-2-deoxy-D-glucose (FDG) continued to re­ceive improvement, particularly to increase radiochemical yield and to remove contamina­tion by the mannose analog. During the latter half of the year routine delivery of quanti­ties of up to 25 mCl of FDG to the hospital were achieved.At the same time the routine production and delivery of 150-labelled water was also main­tained, with back-to-back scans of regional cerebral blood flow (via the water agent) and regional cerebral glucose metabolism (via FDG) being made at a single session for a number of subjects.Work continued on the development of oxygen- labelled molecular oxygen and 1^-labelled77carbon monoxide (required for measurements of regional cerebral oxygen metabolism and re­gional cerebral blood volume, respectively). The problems of encapsulating gases for de­livery to the hospital were not, however, solved this year.Perhaps the most important achievement in the chemistry program in 1984 was the successful completion of the synthesis of 18F-labelled 6-fluorodopa, and its qualification as a PET scanning agent by means of careful demonstra­tion of its chemical purity and of its ste­rility and apyrogenicity, and calculation of patient radiation exposures from measured biodistribution data, as shown in Table XV.Table XV. Biodistribution and radiationdoses for L-6(18F)-fluorodopa.Organ % Activity per gramaDose(mRad/mCi)kBrain 0.25 22Heart 0.72 49Kidney 1.6 284Liver 0.4 65Lung 0.24 31Bladder 0.24 23Skeleton 0.11 34Spleen 0.41 52Ovaries 0.25 47Whole body 44aone hour after injection in rat ^calculated in humanWork began this year on the development of automatic chemical synthesis systems for such scanning agents, with process control by microprocessor. This ultimately will free operator time for the research program and reduce personnel radiation exposure.Delivery of scanning agents between TRIUMF and the Acute Care Unit Hospital continues via the pneumatic transport system (rabbit line) which was commissioned last year. This line has performed essentially faultlessly in this application.The experimental program conducted with the tomograph included both studies of tomograph physics and the application of PET techniques in studies of neurological and psychiatric disorders.Aside from routine measurements of the sta­bility of the tomograph operating character­istics, experiments with 'phantoms' measured the distribution within measured images of events arising from radiation scattering and from random coincidences. It was shown that such distributions are structured (see Fig. 69) as a result of the tomograph scan­ning motion. Studies were also conducted of the propagation (from various measurement sources) of noise into the final measured data on physiological function.Medical research continued on movement dis­orders and dystonia, Alzheimer's disease (senile dementia), schizophrenia, stroke, epilepsy and brain tumours. In the research on movement disorders and Alzheimer's disease the new scanning 6-fluorodopa played a par­ticularly significant role allowing the syn­thesis of dopamine neurotransmitter to be as­sessed in brain cells belonging to the dopa­minergic system, implicated in voluntary movement control and some aspects of demen­tia. The publication listing this year (see p. 171 ) reflects the presentation of many results at scientific meetings.The progress of the research program was largely determined by the availability of irradiation time for the PET target systems, for much of the year only on the CP-42 cyclo­tron. In the last quarter it became possible to produce 18F also from the 500 MeV cyclo­tron, but fluorine production has not yet been possible on both machines simultaneous­ly. Further, beam line 2C from the main TRIUMF cyclotron has not yet produced PET scanning agents, even though it is potential­ly a unique source of several important nuc­lides, including 75Br and 122I. By mid- November of this year a total of 125 research subjects had been scanned, and the scanning rate occasionally approached five and six subjects per week. However, such scanning rates were not maintained for long periods of time, and the synchronization of the logis­tics of production of scanning agents at TRIUMF and the scheduling of subjects to be scanned at the hospital were sometimes very difficult. The work this year on the instal­lation of a switching magnet (with funding from the Woodward Foundation) on the CP-42 system may lead to some alleviation of this problem, via the installation of multiple target systems for PET. Likewise, the simul­taneous availability of PET target irradia­tions in beam lines 1A and 2C of the main machine will eventually also help consider­ably .78POSITION (cm)POSITION (cm )Fig. 69. The effect of gantry wobble motion on the distribution in images of random coinci­dences, measured by a line source of 85Kr (emitter of single photons of 514 keV energy). Projections were measured with the line source surrounded by a cylindrical waterfilled phan­tom of 20 cm diameter. A) Upper: Source located centrally in the phantom which was alsolocated centrally in the gantry aperture; Lower: Source located centrally in the phantom which was in turn displaced 5 cm vertically in the tomograph. B) Source displaced 7.5 cm vertically in the phantom which placed centrally in the tomograph. The saw-tooth contribu­tions to the profiles arise from the to-and-fro motion of the detectors with respect to the radiation source during the wobble motion and lead to ring-artifacts in images.Finally one rather unique event took place this year, namely the marriage of two medi­cally related programs at TRIUMF: PET and the pion radiotherapy project. A subject suffer­ing from astrocytoma (brain tumour), and scheduled for pion radiotherapy in the prog­ram of the B.C. Cancer Institute and Cancer Control Agency of B.C., received a PET scan for regional cerebral glucose metabolism prior to the commencement of the therapy pro­gram. Shortly after the end of one of the subsequent pion irradiation fractions, the patient was returned to the UBC Hospital in time for the positron-emitting debris from the irradiation to be imaged also. They per­mitted delineation of the region of perturbed function caused by the tumour, and also per­mitted confirmation of the accurate steering of the therapeutic dose. It is intended to continue this research with studies of the changes in the glucose metabolic patterns with time after the therapeutic irradiation, and a second subject is currently undergoing scans with this objective.Whole body tomograph developmentThe subject of this research work is the physics development of a novel design of a positron emission tomograph with improvedspatial resolution while reducing the cost and complexity of the instrumentation. Such developments are expected to lead to an ex­pansion of the market for positron tomographs from the area of neurological and psychiatric research to routine clinical diagnostic ap­plications .Prior to starting research as part of the Applied Programs Division, Drs. J.G. Rogers and X.G. Yao had performed computer calcula­tions strongly indicative of the feasibility of the proposed physics developments. A successful grant application was made to the Science Council of B.C. (principal investi­gator B.D. Pate) for a one-year program with­in which a prototype detector system would be constructed and performance characteristics measured, for comparison with the computer calculation results.In the course of the last few months it has become apparent that the commercial implica­tions of this research work extend beyond the building of improved positron emission tomo­graphs. The technology of three-dimensional position encoding of the interaction point of Y-rays with a continuous scintillation detec­tor appears to offer the possibility of radi­cally increasing sensitivity of ordinary79nuclear medicine Y-camera systems, thereby substantially reducing the patient radiation exposures from hospital nuclear medicine procedures. If this turns out to be correct, then we believe that an additional, and very substantial, market exists for modules which could replace many existing hospital Y- cameras to confer the above advantage. In consequence, the patenting of three essential elements of the new developments is being pursued in accordance with the new TRIUMF patent policy.Some of the detailed research progress has been described in a series of TRIUMF design notes [Rogers, TRI-DN-83-39, 83-40, 83-44,84-12, 84-38, and TRI-TN-84-2 (unpublished)]. The earlier general computer study [Yao and Rogers, Nucl. Instrum. Methods (in press)] was applied to the specific case of designing a detector for 511 keV Y-rays which would be suitable for a whole-body PET machine. The new principle employed in the detector is that of three-dimensional position encoding. Conventional PET machines, as well as the more common Y-camera (or Anger camera), mea­sure the positions in two dimensions of Y-rays interacting in a scintillating crys­tal. Because the third dimension of the point of interaction is not measured, both PET machines and Y-cameras suffer from severe compromises in design. In the case of PET machines it is necessary to use an array of long narrow crystals packed tightly together to cover a large area as densely as possible. Such an array is very expensive, difficult to build and maintain, and inefficient when used for detecting Y-rays which originate near the periphery of an extended object like the chest or abdomen.Our idea is to record for each Y-event the position of interaction in all three spatial dimensions. By including a determination of the depth-of-interaction in the scintillator, as well as the usual two transverse dimen­sions, a few large crystals can be used to replace the arrays of small crystals found in conventional ring tomographs. In addition to being less expensive such large crystals should be superior in performance to the small crystal arrays they are replacing be­cause the large crystals are equally sensi­tive to Y-rays impinging at oblique angles to the surface. Efficient detection of such Y-rays is needed to maintain full efficiency at the edges of the field-of-view of a large aperture ring tomograph.Fig. 70. Prototype detector system for second positron emission tomograph.Figure 70 shows a prototype detector system which has been constructed and has been shown to operate correctly in some initial measure­ments . The original objectives of the propo­sal have, however, not been completely met to date. This is due to two factors: a) Theessential large sodium iodide scintillation detector, manufactured by a commercial sup­plier to exacting specifications, has not yet been delivered and is several months late. This delay was in turn due to the failure of a subcontractor to the prime supplier. b) Some of the developments and adjustment pro­cedures respecting the remainder of the de­tector prototype took longer than estimated.(Note added in proof: Spatial resolution mea­surements with a 5 cm thick Nal crystal gave 7 mm FWHM for a collimated 662 keV gamma source.)TRIM PROGRAM123I and 122Xe processing equipment described in the 1983 report has been tested extensive­ly this year, with the result that the moni­toring and control hardware has been shown to80be adequate but the microprocessor is inade­quate under conditions which required more than one task to operate simultaneously. An improved microprocessor has been ordered. The framework of the control program has been established so that programming of the new microprocessor should proceed swiftly when it arrives.Heart-imaging trials with 123i lactone de­rived fatty acids have given excellent re­sults this year in dogs. Canadian and USA patent applications have resulted from this work [A.H. Dougan and D.M. Lyster, Canadian Patent Application Serial No. 436,822, 16September 1983; USA Patent Application Serial No. 647,617, 12 September 1984]. A posterpresentation was also given at the SNM meet­ing in Los Angeles [Dougan et al., J. Nucl. Med. 25_, 122 (1984)]. This work is alsoscheduled for publication in J. Radioanal.and Nucl. Chem. 89, 71 (1985). A simplepreparation was developed for 16-^^I-9-hexa- decenoic acid to secure clinical research which had already been invested in this com­pound. Efficient 123i labelling, ~80%, was attained with phenyl fatty acids. This is better than results reported from most other laboratories. A high performance, liquid chromatograph was commissioned for carrier- free radiopharmaceutical preparations and the associated analyses. Two grants were awarded by the Canadian Heart Foundations to UBC faculty for research concerning fatty acids provided by TRIM.Labelling of brain-imaging 123i amphetamine (IMP) was brought to high efficiency so that IMP can now be offered as a routine pharma­ceutical by VGH using commercial *2 I^. IMP is in demand for the delineation of brain damage soon after the onset of stroke.81CYCLOTRON DIVISIONThe main thrust of the Division during 1984 was geared toward reliable and efficient beam production, and the results achieved were significant. The goal of 300 mAh was reached and exceeded, with a total 322 mAh delivered down beam line 1. Beam was delivered for 5471 h corresponding to 88.5% of the sched­uled operation time, one of the best annual figures achieved until now at TRIUMF, as shown in Fig. 71. In addition, polarized beam operation was improved with delivered current reaching 1 pA at 72% polarization. It should be noted that from May 1984 to year-end the cyclotron was operated without major inter­vention and without requiring a tank lid opening. Although these results could not have been achieved without constant diligent maintenance work by the various groups, a great deal of effort was also dedicated to development.In terms of improved beam parameters the highlight was an extracted cw current of 208 pA tested for over one-half hour through the 1AT2 meson production target. This cur­rent level has not yet been used for produc­tion mainly because the thin 1AT1 meson pro­duction target was incapable of withstanding currents higher than ~140 pA and because of the shorter filament lifetime of the ion source. Induced activation in the cyclotron, excessive background in counting areas and machine reliability also demanded moderation.Improved beam quality and beam performance for beam line 4B proton users became a priority because of the increased difficulty of extracting good quality, low background nA beam currents simultaneously with higher currents on beam line 1 and because of recent extensive use of the MRS spectrometer at small angles and in the dispersion-matching high resolution mode. Specific projects were initiated or received higher priority because of this. These include continuous upgrading of beam phase, magnet and rf voltage stabili­ty, diagnostic equipment allowing the mea­surement of time structure, momentum spread and other quality parameters, high intensity beam phase space definition with slits in the central region, and renewed emphasis on third6 0 -4 0  —mAh-3 0 02 2 5 /j .A  _ (3, 10%20-- 100200m A h / mlOO^iA l2 0 p .A  1 1h /y r6 0 0 04 0 0 02000Fig. 71. Beam charge delivered (broken line) and hours of operation (solid line) over the past several years. Milestones in extracted peak current are also indi­cated. The histogram shows the charge delivered per month.82harmonic flat-topping of the rf waveform. A channel plate system capable of measuring the time structure of the beam within a few hundred picoseconds was developed within the Accelerator Research Division and proved useful for beam quality monitoring. Systems of slits and stripping foils were installed at the entrance of beam line 4 to reduce the spurious background of particles channelled through the beam line, originating mainly from H° atoms produced by electromagnetic or gas stripping of the H“ high intensity beam in the tank. More effort is planned in this direction. Also the third simultaneous­ly extracted beam, the low energy beam on beam line 2C, will receive higher priority. A special task force was set up in collabora­tion with the Applied Programs Division to complete the commissioning of the beam line and corresponding extraction devices in order to achieve a routine isotope production and research station within 1986.There were five major projects active during the year, and three new projects were pro­posed and approved to start during 1985. The resonator improvement program and the altern­ative extraction program were given top priority because of their prime role in reli­ability and future expansion of the machine. At the same time third ion source construc­tion and development of a high intensity op­tically pumped polarized ion source were pursued vigorously. The other active major project, design and construction of the new vertical injection line, progressed within the limits of its budget. Two of the new major projects were proposed to upgrade the cyclotron vacuum system and probes system for reliability and increased efficiency. The remaining new major project, closely con­nected to the resonator program, is a compre­hensive feasibility study of flat-topping the rf waveform in the cyclotron, a highly desir­able feature for high beam quality, high beam intensity and efficient extraction of the high intensity H- beam from the cyclotron.The resonator improvement program progressed favourably, the highlight being a substan­tially upgraded cyclotron rf diagnostics system to allow deeper understanding of phe­nomena affecting reliability of the existing resonators. A new prototype resonator seg­ment will not only improve alignment but also will be compatible with very low vibration amplitude at the tips (~±2 pm), required for rf voltage stability and stable third harmon­ic flat-topping. A segment with improved structural features assembled and measured inthe laboratory provided a factor of three in­crease in tip rigidity and about a factor of 10 reduction in tip vibration amplitude. This prototype was tested at power in the rf vacuum test stand in the laboratory and will be installed in the cyclotron during the February 1985 shutdown to verify its compati­bility with the rf leakage inside the cyclo­tron tank.Third harmonic excitation of the cyclotron resonator cavity was tested at half power in air during the spring shutdown. The test showed the dee gap voltage of the third har­monic component to be decreasing from the centre of the machine toward the outside by about 80%. Studies were carried out on the 1:10 model and with a special program derived from an RFQ code developed at Chalk River. It was shown that the voltage drop can be corrected by altering the capacitance in the centre and in the flux guide regions. 1:10 model studies to simulate third harmonic control mechanisms are proceeding and giving encouraging results. One of the most sensi­tive aspects is the degree of stability needed for both third and fundamental har­monics. This requires not only a more rigid and stable mechanical configuration but also an accurate rf control network which will correct for the effect of residual mechanical instabilities. A special control system is being commissioned on a two-segment cavity test stand where fundamental and third har­monic have both been tested at power and a flat-top waveform has been achieved. Studies to anticipate difficulties encountered when stability control is in the 80 resonator segment cyclotron system are being carried out, both through the 1:10 scale model and through the computer code.Although H- extraction studies were only funded in April, significant progress had been realized by year-end, especially in concept. A new method to increase the separa­tion between turns at extraction, based on driving the vr = 1.5 resonance by means of an 11.5 MHz rf radial deflector to produce coherent radial oscillations, was proposed by a task force involving the Accelerator Re­search and the Cyclotron Divisions. By year- end the essential elements for demonstrating an extracted orbit separated from the intern­al beam were in an advanced stage of design and prototyping. A prototype rf deflector was prepared for installation in the cyclotron during the February 1985 shutdown. Also a short prototype of the first electrostatic deflector was constructed to evaluate the83effects of magnetic, thermal and radiation fields in the cyclotron. Special attention was paid to the design of the 1 m long anti­septum and its insulators operating at a positive voltage of 60 kV. Preliminary stud­ies on an iron-free magnetic channel with minimal perturbation of the inner orbits were completed. A cavity to boost the energy gain in the region of 450-500 MeV to increase the spacing between successive turns and reduce the number of turns at radii where electro­magnetic stripping becomes important was de­signed and a prototype constructed. Tests of prototype components with a pulsed beam will proceed during the next three shutdowns. As the tests prove satisfactory, prototypes will be replaced by permanent devices. So far there is no reason to believe that a high in­tensity H- beam cannot be extracted reliably from the cyclotron.Progress on the injection line and ionsources was considerable. The third ionsource terminal was erected before year-end, the transformers and 300 keV power supply were installed and the system was ready to be tested at full voltage. Design of the ion source interlock logic and of the connecting injection line was complete and several com­ponents were constructed and assembled. The terminal is large enough to contain twosources. It is planned to install first a high-intensity high-brightness ion source which will likely be an H“ CUSP source. This will be followed by installation of the optically pumped polarized H~ ion source, now under feasibility study in the laboratory. The terminal has been built with flexibility and size so that ion sources can be easily interchanged with other types, as required by utilization and development of the machine. Commissioning of the third terminal will improve reliability of the ion sources, both polarized and unpolarized, and allow develop­ment toward higher extracted beam intensities (up to 400 pA unpolarized and 10 pA pola­rized). Extensive operation at 400 pA cur­rents will depend on the success of thebooster cavities in the cyclotron since these are expected to reduce beam loss due to elec­tromagnetic stripping by about a factor of three with equivalent reduction in radiation build-up.Funds were allocated in the vertical injec­tion line project for design and installation of a third buncher to enhance the bunchingaction of the first and second harmonicbunchers previously installed. This element is required to reduce energy dispersion in­troduced by space charge in the bunched beam. Also, the third buncher will be instrumental in providing a very high concentration of beam in short (±6°) phase-space intervals re­quired by the kaon factory design. By year- end the third buncher was designed, built and ready for installation in the vertical injec­tion line during the February 1985 shutdown.The optically pumped polarized ion source was completely assembled in the laboratory and several technical difficulties related to operating the system in a cw mode were over­come. The 28 GHz ECR proton source was ope­rating at predicted intensity levels. Laser polarization of sodium in the vapour cell was very successful. Using a viton liner on the cell walls, the lifetime of polarized atoms could be increased by a factor of 10 compared to that for metal walls. The formation of the polarized beam was indirectly demonstrated and the commissioning of the source toward higher intensities and better beam quality is now in progress. One should note that this effort was carried out in collaboration with colleagues from KEK, Los Alamos and LBL and is putting TRIUMF in the forefront of H- polarized ion source research.Although development activities of the Vacuum and Probes groups had not yet been organized as major projects, important steps were taken toward essential improvements in both areas. Operational vacuum levels around 1 or 2xl0-8 torr were achieved in the cyclotron. A spe­cial turbomolecular pump and a special cryo- pump were tested, ready for installation in the tank to verify adequacy in the cyclotron environment. The aim is to replace existing diffusion pumps, which present the danger of oil contamination, have inadequate pumping speed for hydrogen, and are costly to operate.The Probes group designed and tested a new low-energy probe in which most of the track and moving system are grounded and shielded from the stray rf field. The new probe is complete, tested in the laboratory and ready for testing in the cyclotron tank.Some minor projects should also be mentioned. For instance, the new remotely controlled periscope viewing system, developed by the Diagnostics and Engineering Physics group, was a great step forward in viewing and in­specting the cyclotron tank with no radiation exposure to personnel. The Remote Handling group on several occasions demonstrated its capability of dealing with increased radia­84tion levels, for instance with more efficient remote installation of protective lead shielding around the tank periphery during shutdowns, with low dose repairs of quadru- poles and replacement of vacuum seals, espe­cially around the 1AT1 and 1AT2 targets, and with the remote replacement of resonator seg­ments in the tank. A special platform was set up in the remote handling building to test the installation and removal of new resonator segments in connection with the resonator improvement program.A deficiency in the low conductivity water cooling system, which in the past had caused contamination of the beam line magnet cir­cuits and some corrosion of cooling lines, was corrected. Guy LeDallic, on leave from CERN, was in charge of this project and designed and implemented a series of improve­ments which produced, with other ameliora­tions, factors of five to 10 in water resistivity.On the controls side the most significant improvement in the control room was the addi­tion of a VAX 730 to serve the cyclotron Operation and Development groups in their analysis of the data available through the existing system. The new VAX is connected to the control CAMAC system and to the TRIUMF central computer (VAX 780) via an ETHERNET network link. This system has already proved helpful in the reproduction of cyclotron beam tunes and in the monitoring and control of machine isochronism and of rf voltage and other machine parameters vs. time.Factors limiting development were the limited access to equipment due to extended beam pro­duction periods and radiation exposure to personnel during interventions. With shield protection installed around the tank periphe­ry the radiation levels in the tank are now around 70 mrem/h at the centre and around 300 mrem/h at the periphery. These figures agree with what had been predicted a few years ago, for the present level of about 300 mAh of total accelerated current per year. The dose to personnel has reached in average a total of about 60 rem per year, distributed over about 120 people. The dis­tribution sees, obviously, some people re­ceiving doses close to the 1 rem per year limit, which is set by TRIUMF guidelines, and most people with much lower doses. Improve­ments in remote handling, reliability and efficiency of the cyclotron tank instrumenta­tion and the insertion of the rf boosters in the cyclotron are expected to reduce exposurelevels in the future and may allow higher levels of available current for the meson users.BEAM PRODUCTIONOnce again beam production records were set in 1984. The total integrated current for the year was 322,373 pAh, a 33% increase over1983 which was our previous best year. This is a result of routine operation of 130 to 150 pA extracted in beam line 1A during the summer. Figure 72 shows the unpolarized beam production for the year, which was 80% of the scheduled production.During the summer (schedule 56) two weeks had beam production of greater than 17,000 pAh and three weeks had beam production at 100% or greater than what had been scheduled be­cause of cyclotron development work towards higher currents. On July 25 a 500 MeV proton beam current of 208 pA cw was extracted con­tinuously for 1/2 h following a number of de­velopment shifts at high equivalent currents.Similar currents were scheduled for the fall (schedule 57) but 140 pA was not achieved un­til the last three weeks of unpolarized ope­ration. Currents were limited by the condi­tioning of a new TNF protect monitor (1AM11) and the new TNF window (1AWZ) plus the com­missioning of the radiogas target of TNF. All these items were installed in the September shutdown. Also a limitation on the current was the thermal damage to the 1A targets caused by high beam densities and the lack of diagnostics to measure beam densities.Figure 73 shows the operating hours for 1984. The hours of operation include development, cyclotron tuning and beam to experiments for a total of 5471 h, which is 88.5% of the scheduled operation. The hours of unpolarized beam on target in beam line 1A for experi­ments were 2904 h compared with 2932 h for 1983. Polarized beam was scheduled after each high current run for a total of 1046 h, about 8 weeks, which is a significant in­crease over 843 h for 1983. Polarized beam was 70% available.Figure 74 shows the operating record for 1984 and includes the downtime by group. Downtime which is lost time due to equipment failure has decreased from 719 h in 1983 to 540 h in1984 and is down significantly in RF and ISIS, including P0LISIS. The only failure which required rescheduling and does not85£  iooooI ■ i ■ ■ I i i i i I i i i » I i i i i 1 ■ i » i L i-L -L jJ  L U jlt iX L L u X lli404,228 ufl Hrs. Scheduled / —322.373 ufl Hrs. Delivered  J/W e e kFig. 72. 1984 beam delivery.We e kFig. 73. 1984 hours of operation.appear as downtime was a water leak in 1AQ16, which required four days to repair in weeks 45 and 46 (schedule 57). A water leak in 1AQ12 and an M8 vacuum leak were also re­paired in schedule 57. The only other signif­icant downtime was the failure of the 200 hp pump for the active aluminum water system, which caused two days downtime during the summer. The response of all groups to equip­ment failure was exceptional and should be cited as one of the main reasons for the low downtimes.Two weeks of beam production were lost in May during start-up following the spring shut­down. Some of the tank diagnostics and cor­rection plates were damaged by third harmonic tests and the rf dee voltage was misinter­preted because of new voltage probes instal­lation, making it difficult to reproduce the beam at start-up.Table XVI shows the scheduled and delivered beam to each experiment during the year. A 'P' in the hours column indicates polarized operation.A brief summary of the year's operation is:Cyclotron ONBeam to experimentersDevelopmentTuningCyclotron OFFShutdownMaintenanceDowntimeOverhead, start-up, etc.4565 h 463 h 443 h5471 h1954 h 428 h 540 h 345 h3267 hMAINTENANCE ( 428)OTHER ( 45)SHUTDOWN ( 1954)STARTUP (100)DOWNTIME (540)OVERHEAD ( 200)BERNOPERATION (54711DOWNTIMEIBFEKtor hiW ES It!OTHER (9)TARGETS (9)TRIPS 129)DIAGNOSTICS I 35)VACUUM (37)BERHLINES 149)Fig. 74. Operating record for 1984.CYCLOTRONCyclotron developmentDuring 1984 cyclotron development activities were directed mainly to the study of the TRIUMF rf accelerating system. Considerable effort was directed towards experimental, theoretical and model studies of the resonat­or structure, with the overall aim of achiev­ing a flat-topped accelerating waveform, im­proving the voltage stability, and reducing the leakage. Efforts also continued towards improving the reliability of the machine in terms of total delivered charge, and towards the improvement of the quality of the beam in terms of reduced energy spread and reproduci­bility of tunes. Particular attention was86Table XVI. Beam to experiments - total 1984.Experiment* Channel Scheduled Deliveredh yAh h yAhieson Hall:Test M9 46.0 6440.0 44.7 9057.7B M13 37.5 4875.0 37.3 4410.8KC Mil 33.0 990.0 30.4 929.8KR test M20 23.0 690.0 17.4 798.0Test M15 81.0 11340.0 41.5 3233.4Tune Mil 81.0 + 439.0 P 2430.0 91.9 + 304.9 P 2052.7Spin rotation M20 47.0 4700.0 27.7 575.0QQD development Mil 127.0 4410.0 42.7 233.9QQD Mil 418.8 54437.5 397.8 45219.4QQD M13 763.0 94510.0 693.8 87461.047/tune M20 81.0 2430.0 57.4 1841.7104 M9 3384.5 389385.0 2842.8 301607.6147 M15 46.0 6440.0 39.5 4176.9147 M20 660.0 64130.0 537.8 51692.2150 M20 393.0 44290.0 296.6 28919.7157 M20 69.0 9660.0 70.8 9038.4161 M20 127.0 16510.0 123.9 15285.0191 M20 127.0 14080.0 117.2 11902.4204 M13 150.0 21000.0 98.9 11838.4205 Mil 1006.0 104440.0 752.9 67641.9217 M13 948.3 96752.5 688.9 59702.2219 Mil 427.0 55510.0 371.1 48642.3230 M20 162.5 21125.0 163.3 22311.9231 M9 59.3 7702.5 60.2 7401.9232 M20 334.0 44690.0 214.8 26252.0233 Mil 46.0 5980.0 47.2 6305.8239 M20 196.0 24100.0 192.1 21265.3241/157 M20 172.0 9220.0 150.7 6604.6243 Mil 346.0 44980.0 333.8 41224.7244 M15 46.0 6440.0 44.8 5976.1244 M20 104.0 13520.0 104.6 13297.0244/5 M15 46.0 6440.0 48.3 5918.8245 M15 58.0 8120.0 56.5 7527.7245 M20 104.0 13520.0 99.9 11118.2246 M13 335.0 46900.0 318.9 39351.6247 M13 518.0 51770.0 474.5 44794.5248 M13 473.0 + 127.0 P 46200.0 447.7 + 79.6 P 44294.3249 test M9 58.0 1740.0 33.0 1172.5250 M13 173.0 22490.0 157.5 19199.6254 Mil 139.0 13900.0 112.3 10827.6255 M13 127.0 17780.0 56.4 5657.6260 M20 266.0 35850.0 244.1 30707.6261 M15 92.0 12880.0 87.1 10226.2262 M20 265.0 29620.0 208.2 22848.0263 Mil 554.0 70690.0 522.3 65057.9270 Mil 150.0 21000.0 98.9 11838.4273 M20 23.0 2990.0 24.0 3132.3275 Mil 150.0 19500.0 145.7 18644.3275 M20 105.3 13682.5 107.6 13440.7286 M20 35.0 4900.0 30.5 4465.4288 M15 46.0 6440.0 41.5 5378.3290 M20 254.0 35560.0 202.2 23743.6291 M15 46.0 6440.0 6.5 205.887Table XVI (cont'd)Experiment* ChannelhScheduledpAh hDelivered296 M15 104.0 14560.0 103.3325 M15 35.0 4900.0 11.8787 Mil 70.0 7000.0 49.5208 IB 980.0 P 677.0 PProton Hall:Perm, magnet 4A 81.0 33.2218 4A 150.0 131.0121 4A/2 58.0 + 230.0 P 30.6 + 176.5 P121/190 4A/2 173.5 + 774.0 P 116.1 + 629.2 PMRS 4B 987.5 + 104.0 P 483.4 + 92.3 P(p,n)(n,p) test 4B 104.0 80.3142 4B 277.0 204.2169 4B 127.0 114.1206 4B 403.0 245.2212 4B 254.0 188.6215 4B 127.0 70.4216 4B 69.0 27.8218 4B 69.0 53.9221 4B 69.0 P 34.3 P223 4B 254.0 149.9227 4B 127.0 92.2234 4B 127.0 49.6236 4B 104.0 P 73.6 P238 4B 313.0 157.1pAh11905.21612.2767.4*See Appendix C for experiment title and spokesman.paid to high split ratio simultaneous extrac­tion modes and to the development of new diagnostic techniques.r f studiesFor the upgrading of beam stability, beam quality and intensity, and for future devel­opments, it is necessary to achieve and reli­ably maintain a flat-topped waveform in the accelerating rf field. Because of the low value of the accelerating voltage, resulting in a large number of closely spaced turns in the machine, the stability requirements on the amplitudes and relative phase of the fundamental and third harmonic voltages are extremely strict.The computer code RFQ3D obtained from CRNL was implemented on the VAX-11/730 computer. The program, although written to study rf quadrupole fields, was adapted to calculate voltage profiles along the TRIUMF dees bytreating the rf resonant cavity as an rf dipole. Cavity perturbations, including the centre post, flux guides and rf panel mis­alignments have been successfully simulated in the code and their effects on the voltage profile evaluated. This was previously im­possible with any known 3-D code.The measurement of the effective or acceler­ating dee-to-dee voltage, as seen by the beam, at all radii, is of fundamental import­ance. To obtain an approximate measure of the dee voltage profile capacitive rf voltage probes were designed, constructed, tested and installed in the ground arms immediately opposite the hot arm tips. Because of their nature these probes measure only the electric field generated between the hot arm tip and the ground arm. If the resonator tip sags or vibrates, this will effectively change the calibration of the voltage measurement. A total of 60 voltage probes were installed along the dee tips in collaboration with the88RF group. The probes and the semi-rigid phase stabilized coaxial cables connecting them to the control system have been careful­ly calibrated in the rf test stand before installation and have achieved a sensitivity of 0.15 dB.Fundamental and third harmonic voltage pro­files along the dee gap were measured in a 1:10 scale model of the resonator cavity under a variety of conditions. Figure 75 shows the calculated versus measured voltage profiles, for both 23 and 69 MHz. The re­sults show that even with perfect alignment of the rf panels, the voltages are not uni­form. The calculations are in good agreement with model results and with measurements made in the cyclotron. The increased tip-to- ground capacitance near the centre post re­duces the resonant frequency and subsequently increases the voltage in this region. The effect is reversed in the flux guides, where the effective tip-to-ground capacitance is reduced. In the machine the third harmonic voltage measured at the outer segments is 10 dB lower than that measured at the centre segments. This ratio has been reduced, in the 1:10 model, to approximately 4 dB simply by installing electrical shorts in the root region of the centre segments to counteract the effect of the tip capacitance. After correcting for the centre post numerical calculations show that the flux guides also cause detuning. By optimizing these guides to resonate at the frequency of the rest of the cavity, variations in the voltages of less than 1 dB were obtained.Numerical calculations and measurements show that if an upper resonator tip is dispersed, the voltage in the lower segment in the sameFig. 75. A comparison of rf dee gap voltage profile measurements and RFQ3D calculations.quadrant will be perturbed due to capacitive coupling between opposing dees. The cavity Q will also be altered resulting in changes in the power delivered to the cavity, conse­quently altering the hot arm voltages. Volt­age regulation will tend to reduce this per­turbation. The average accelerating field, as measured using the beam time of flight, was the most stable when the probe signals from two opposing quadrants were summed, thereby more closely approximating the volt­age seen by the beam. Changes in the natural resonant frequency of the resonator due to vibrations of the hot arms produce fluctua­tions in the relative phase between the tip and the coupling loop of as much as 5° rms. However, very good relative phase stability is measured between the segments of a given dee, due to the strong magnetic flux coupling within a given dee.An improved version of 2-d rf cavity code SUPERFISH was obtained from LANL and imple­mented. This version can handle regions of different permeability as well as different permittivity. The energy and cavity Q calcu­lations are separated from the main core of the iteration routine, thereby eliminating some of the previous problems encountered with unusual geometries.Because of the tolerances required for reso­nator installation and alignment a survey of the outside surface of the cyclotron tank lids (top and bottom) on which the resonators are mounted has been performed. A simple, precise technique using two connected gradu­ated cylinders filled with water was used. Both the lid and bottom were measured to be flat and horizontal to within ±2 mm, indicat­ing that no major base misalignment exists in the cyclotron tank. A comparison of these measurements with periscope data, which are more suited for relative comparisons, is shown in Fig. 76.Beam phase stability and tune reproducibilityBeam phase stability of better than ±1° in beam line 4B has been demonstrated for peri­ods of several hours. The technique uses microchannel plates (developed by the Beam Line Diagnostics group in the Accelerator Research Division) to sample all of the beam phases. A digital signal averager (centroid determination) module has been developed in which pulse derived sampling signals can be prescreened and averaged to produce a running analog signal. The device is capable of ac­cepting input rates of up to 125 kHz and can89DISTANCE FROM TANK CENTRE [inches]Fig. 76. Periscope and water level measure­ments of the flatness of the cyclotron tank bottom.compute consecutive averages of from 1 to 32,000 successive samples. The digital win­dow can selectably 'follow' the data drifts. The device is used to select and average over a portion of the phases. The required phase stability of better than ±2° has been achieved under the 'worst case conditions',i.e. 120 pA in beam line 1A, 1 nA in beamline 4B (split ratio <1:100,000), both at 500 MeV extracted energy thereby accumulating the maximum phase noise. The phase profiles obtained with the channel plates have been demonstrated to be useful as a diagnostic and tuning method for beam production down beam line 4.DiagnosticsA prototype split-plate, non-intercepting beam position monitor has been constructed and tested in the injection beam line. The device works on the 1 kHz beam pulse struc­ture and uses a gated dc restore circuit to recover the signal from the noise due to mechanical vibration and 60 Hz pickup. The position accuracy of the device is ±0.1 mm. A system to measure the extracted beam inten­sity using the second capacitive probe on beam line 1 has also been successfully tested. This measurement uses the 23 MHz microstructure of the beam, and is not depen­dent on the 8 ys ISIS hole in the beam, which is sometimes filled in by poor cyclotron tunes.Data acquisitionThe hardware interface between the VAX 11/730 and the central control system CAMAC execu­tive system has been installed and tested, and has been operational for several months. The software interface between users of the VAX VMS operating system and the cyclotron is tested and complete. Read access to any cyc­lotron parameter is permitted for any user, whereas write operations can only be carried out by users present in the immediate area of the cyclotron or the control room. A CAMAC- to-CAMAC communication link has been estab­lished, providing for high speed transfer of data between any of the central control sys­tem computers and the VAX. The 730 is con­nected to the main VAX-780 via a DECNET- ETHERNET link, allowing software developed on the 780 to be executed directly on the 730. This has provided immediate availability of very powerful graphics and analysis packages on the 730.A number of application programs have been developed to assist the operators in machine tuning:• V*cos(<(>). This program takes successive readings of the time of flight of the beam between injection and arrival at the HE2 probe, converts and plots the data in the form of the cosine of the beam phase multi­plied by the local accelerating voltage vs. the radius. This provides information on the isochronization in the cyclotron. The data are digitally filtered to reduce the noise in the measurement. The V*cos (<j>) curves are stored on disc for future reference.• Centring scans. These scans are used to determine how well the beam is centred in the machine, by measuring consecutive inter-turn separations over an extended region. These data are used to adjust the appropriate har­monic coils to minimize the centring error.• An interactive logging and analysis package has been written for the VAX. The user can select any number of parameters for measure­ment and storage at selectable time inter­vals. The package has been useful in perform­ing correlation studies between machine and beam parameters over extended time periods.A serial CAMAC highway now connects the VAX to CAMAC crates in the control room, the RF room, ISIS, level 264, the 1:10 model, the vacuum test facility and the rf test stand.90The device driver software supports pro­grammed I/O, DMA block mode transfers (Q-stop and Q-scan) and list mode transfers, in which arbitrary lists of CAMAC calls including var­iable length block transfers are carried out entirely by the hardware without requiring the VAX CPU. Software for carrying out ex­periments on the 1:10 model, the vacuum test stand and the rf test stand is operational.RF systemResonator improvement programAfter testing of a new resonator segment had been completed in 1983 and some improvements were dictated because of revised specifica­tions, the highest priority in the resonator program became the design, manufacture and actual installation in the cyclotron tank of an improved, highly stable prototype reso­nator segment by February 1985. Improvements were intoduced to the hot arm in the follow­ing areas:1) Dynamic stability. The need for increased voltage stability and the potential require­ment for third harmonic flat-topping demand a stable hot arm with lower tip vibration amp­litude (possibly 10x). To ensure a high degree of dynamic tip stability water flow excitation due to turbulence in the rf panel requires significant reduction.2) Long-term stability. Temperature and stress levels require control to avoid defor­mation or creep. Material selection should be consistent with these needs.3) Profile. It is important to have and maintain a straight and flat hot arm such that asymmetries between upper and lower segments are avoided. RF leakage into the beam gap causes strongback heating and tip instability which currently is compensated for by an exact tip alignment procedure. A design less sensitive to heat absorption would allow a greater control of leakage hence more stable operation.In order to assist in the design phase a development test program was initiated to examine the dynamic stability (items 1, 3above). The test stand in the proton hall extension was utilized and measurements were also taken on the existing resonator panels during the March shutdown. A study was done in conjunction with a specialist from UBC. The findings of this investigation are pub­lished. In brief, the recommendations were:1) To maximize structural damping2) To maximize the fundamental frequency of the hot arm3) To consider the use of stainless steel as a candidate structural material4) To use an rf panel configuration employing parallel longitudinal water flow patterns instead of the transverse serpentine pat­tern used in the existing resonatorsThermal stability requirements indicated the need for water cooling on the beam side of the resonator segments. A structure with a water-cooled strongback was envisaged in the design with the possibility of addition of a water-cooled strongback cover, should this be required due to excessive local heating.In July a design specification and schedule were created and detail commenced. This work was done with assistance from Canadian Air­craft Products. The installation of one (possibly several) segment in the present resonator necessitated that the new design be compatible with existing adjacent segments, support structure, water manifolding and remote handling equipment. As a result design changes focused only on the hot arm. The original levelling arm, root, ground arm and tip were reused with minor changes. Auste- nitic stainless steel was ruled out in favour of aluminum as a structural material due to evidence of a slight increase in magnetic permeability in the cyclotron environment.The improved strongback design employs four extruded 'I' beams with integral flow chan­nels riveted to an upper and lower cover plate (AH 2024 - T 351) that tapers from root to tip in steps. A special assembly fixture was designed with 10 contour clamping sta­tions such that a controlled curvature could be built into the strongback ensuring a flat profile when cantilevered from the normal supports. The rf panel is constructed from 0.032 in. Cu sheet with 304SS tubes soldered to the panel. The tubes are flattened for low profile, high contact area. The pattern of the tubes consists of 14 parallel paths from root to tip compared with the original roll bond panel that has 3 paths in series in a serpentine pattern.Assembly of the new hot arm was completed in early December and testing commenced at that time on the test fixture. The following re­sults were achieved, shown compared to those of an original hot arm:91Original PrototypeTip stiffness (lb/in.) 55 149 (calc)143 (meas)Hot arm weight (lb) 456 559Fundamental frequency (Hz) 4.75 5.9 (calc)5.4 (meas)Tip vibration amplitudewith no damper as measuredin the lab (in. pk/pk) 0.005 0.00005Tip vibration amplitudewith damper 0.001The new hot arm will be assembled to its root and ground arm and installed into the rf test facility for thermal testing early in the new year. At the same time a second prototype isbeing assembled with a corrected strongbackprofile for flatness (within ±0.040 in.).This unit will be assembled into the cyclo­tron in February 1985.Third harmonic test facilityThe rf test facility in which two segments are assembled under vacuum to form a 23 MHz cavity was successfully operated under high power with a flat-topped rf voltage of 70 kV just before year-end. Work began immediately thereafter toward commissioning the elements for automatic tuning plus voltage and phase control when operating with the flat-topped accelerating voltage. This represents a sig­nificant milestone toward the achievement of a third harmonic flat-topped waveform in the cyclotron, since multipactoring, stability and control feedbacks can be tested at full power in the test stand before being incorp­orated in the final design.Multipactoring proved to be no problem in operation under vacuum. The system was turned on at 23 MHz by pulsing. Once the 23 MHz voltage was on continuously above the multi­pactoring level the 69 MHz voltage could be turned on from low levels independent of rise time.In the development of the system leading up to high power operation the following tasks were completed:1) A three-stub tuner was built for impedance matching of the third harmonic loop. The return loss measured less than -18 dB over the tuning range of the cavity. The fundamen­tal loop was found to be critically coupled over the tuning range of the cavity with a return loss of <-20 dB. Measurements at sig­nal level indicated that the amount of funda­mental power coupled back into the third harmonic transmission line was down by -45 dB (1 W for input of 40 kW). The amount of third harmonic power coupling back into the funda­mental transmission line was down by only -3 dB (2500 W for input of 5000 W). A first attempt to handle this problem was made by reconfiguring the 3-element matching network for the fundamental loop as a low pass filter. This improved the isolation from-3 dB to -11 dB (400 W for input of 5000 W). Different filtering schemes with greater at­tenuation are being investigated.2) A digital motor-servo system was completed to drive the tip and the swayback mechanism for tuning.3) Phase detectors and filters were designed and tests begun for use in automatic tuning of the fundamental and third harmonic.4) The 69 MHz power amplifier as originally designed and built by a local firm was found to be unsatisfactory. The amplifier was re­designed and built in collaboration with M. Maerki of SIN. Operation to 16 kW output in a dummy load was achieved.5) The control system for generating the fundamental and third harmonic signals and controlling both voltage and phase signals was completed, bench tested, and used for initial tests on the test facility resonator. Tie-in to a VAX 730 computer was made for initial data-handling and first attempts to control the overall system.Operational performanceThe resonator has continued to operate with no indication of sagging of the tips.As had happened in one earlier case, a bel­lows in a water-cooling line (segment Q2U4) developed a water leak tha