UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Gamma radiations from the bombardment of boron ten with protons and deuterons Chadwick, George Brierley 1955

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

Item Metadata


831-UBC_1955_A8 C4 G2.pdf [ 3.39MB ]
JSON: 831-1.0085328.json
JSON-LD: 831-1.0085328-ld.json
RDF/XML (Pretty): 831-1.0085328-rdf.xml
RDF/JSON: 831-1.0085328-rdf.json
Turtle: 831-1.0085328-turtle.txt
N-Triples: 831-1.0085328-rdf-ntriples.txt
Original Record: 831-1.0085328-source.json
Full Text

Full Text

GAMMA RADIATIONS FROM THE BOMBARDMENT OF BORON TEN WITH PROTONS AND DEUTERONS by George Brierley Chadwick A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in the Department of PHYSICS We accept this thesis as conforming to the standard required from candidates for the degree of MASTER OF ARTS Members of the Department of Physics THE UNIVERSITY OF BRITISH COLUMBIA August, 1955 ABSTRACT The gamma rays resulting from the bombardment of with protons of energies from 0.5 to 2.0 Mev. have been observed with a sodium iodide s c i n t i l l a t i o n counter. Capture radiation, of energy Eg = 8.81 ± 0.05 + 10/11 E p Mev., showed a broad resonance at Ep = 1135 1 15 kev. At this energy, the radiation had an angular distribution of the form 1 + (0.5 * 0.05)cos2© and a total cross section (3.5 * 1.0) x -^0 2 10 J cm . Several lower energy radiations were also observed and assigned tentatively to cascade transitions i n e n . The cross section for the M-30 kev. radiation from the reaction B 1 0(p,ar)Be^ was measured to be 0.21 *" 0.05 bn. at Ep = 1.52 Mev. This radiation was found to be isotropic. A three crystal s c i n t i l l a t i o n counter pair spectometer has been constructed and used to observe the gamma radiations from the bombardment of B-^ with deuterons of energies from 0.6 to 2.0 Mev. The spectrometer incorporated a selection of those pulses i n the side counters representing 0.5 Mev. energy release before they entered a tri p l e coincidence c i r c u i t , responding to coincidence between the three counters. The improved energy resolution and background rejection allowed the complex gamma ray spectra from B ^ + d to be resolved and interpreted. 0 ACKNOWLEDGEMENTS The author wishes to express his gratitude to Dr. J. B. Warren for his attentive supervision of the work described i n this thesis. He i s also indebted to. his collaborators: Mr. T. K. Alexander for his help with the proton bombardment work, and Messrs. J. T. Sample and G. C. Neilson for their co-operation i n the deuteron bombardment studies. Thanks are due as well to Doctors D. B, James and K, L. Erdman for their able assistance i n developing the three crystal spectrometer. Messrs. K. A. Laurie, G. Jones and J. B. E l l i o t t deserve thanks for their willingness to help with the runs. Finally, the receipt of two scholarships from the National Research Council of Canada i s gratefully acknowledged. TABLE OF CONTENTS Chapter T i t l e . Page I , INTRODUCTION 1 II DETECTION OF THE RADIATIONS FROM.THE PROTON BOMBARDMENT OF BORON TEN ... 5 1. History and Expectations 5 2. Apparatus 7 (a) The Target 7 (b) The S c i n t i l l a t i o n Counter .... 8 (c) E l e c t r o n i c s 11 3. Experimental Procedure 12 k. Results 1*+ (a) The 9 Mev. Radiation ih (b) The Intermediate Energy Radiation 20 (c) The Low Energy Radiation .... 20 5. Discussion , 21 (a) The Width of the Gamma Ray Tra n s i t i o n 21 (b) The Spin of the Excited State of C 1 1 23 I I I THE THREE CRYSTAL PAIR SPECTROMETER .. 28 1. P r i n c i p l e of the Spectrometer ... 28 2. Details of the Spectrometer 32 3. Operation and Results 37 IV GAMMA RADIATION FROM THE DEUTERON BOMBARDMENT OF BORON TEN * f l APPENDIX |2 REFERENCES LIST OF ILLUSTRATIONS Number Subject Facing page Plates I. The three c r y s t a l spectrometer apparatus 37 I I . The three c r y s t a l counters 37 Figures 1. Energy l e v e l s i n the nucleus C l l 5 2. C i r c u i t diagram of the s c i n t i l l a t i o n counter head amplifier and power supply 11 3 . Pulse height d i s t r i b u t i o n from the 9 Mev. radiations from the reaction Bl°(p,y)C"ll Ih h. E x c i t a t i o n function of the 9 Mev. r a d i a t i o n 16 5. Angular d i s t r i b u t i o n of the 9 Mev. r a d i a t i o n 17 6. Intermediate energy r a d i a t i o n from Bl° + p 20 7. Lower energy radiations from Bl° * p * 21 8. Single c r y s t a l spectrum of the radiations from Bio + d 28 9. Block diagram of the three c r y s t a l pair spectrometer 29 10. D i f f e r e n t i a l discriminator c i r c u i t diagram ....... 33 11. Anticoincidence and t r i p l e coincidende c i r c u i t s for the side channels 33 12. Gate pulse generator c i r c u i t 3^ 13. Gated biased amplifier c i r c u i t 35 1*+. Three c r y s t a l spectrum of the radiations from the reaction F19(p,«ar)ol6 39 15. Single and three c r y s t a l spectra of ThC" r a d i a t i o n . 1+0 16. Three c r y s t a l spectrum of the radiations from Bl° + d, E, - 1.1+ Mev., 2 - 5 Mev. region 17. Three c r y s t a l spectrum of B 1 0-»-d, h - 7 Mev. region hi 18. Three c r y s t a l spectrum of Bjj-0 +d, 5 r 8 Mev. region hi 19. Three c r y s t a l spectrum of Bl° •*• d, 6.5-9 Mev. region i+l Tables I. Resolution of the s c i n t i l l a t i o n counter I I . Angular d i s t r i b u t i o n c o e f f i c i e n t s for pure multipole r a d i a t i o n . . . ;V.:. 11 26 CHAPTER I INTRODUCTION The fact that most nuclear emanations are mono-energetic has been taken to signify that the effects of nuclear forces may be dftscribed by a quantum theory of the nucleus. Various force laws, a l l implying that the elementary particles forming the nucleus"maintain a separate identity, have been used with at least qualitative success to explain the results of experiments. Unfortunately, the many body problem involved i s so complex mathematically that i t has been to date impossible to judge whether quantitative agreement could be obtained i f accurate calculations were possible, or whether the whole approach i s i n error, and qualitative agreements with experiment mainly fortuitous. The shell model of the nucleus has been eminently successful i n i t s qualitative predictions while avoiding the actual solution.of the many body problem. The theory based on this model proposes that fundamental particles i n the nucleus, obeying the Pauli exclusion principle, " f i l l up" states of differing quantum numbers i n the average f i e l d of the nucleus. Spin-orbit and spin-spin coupling are postu-lated to account for the observed sequence i n energy of these states. The success of this model i n predicting ground state angular momenta of nuclei has made the level structure of nuclei a subject of intenswe study, particularly i n those nuclei of mass number less than twenty. Nuclei may be formed - 2 -in excited states by bombarding them with fast particles; the decay of an excited nucleus can produce emanations from whose energy and angular dependence of intensity the energy,y A 11 spin", and "parity" of the excited nucleus may be deduced. Ajzenberg and Laritsen (1955) have compiled the results of such experiments i n a review a r t i c l e . Experimentally, these quantities may be most easily and precisely measured by observation of the charged particles emitted i n nuclear reactions. However, radiative transitions bear at least a formal similarity to optical transitions, and should be more easily interpreted i n terms of nuclear models than particle emission with no non-nuclear analogy at a l l . For example, Weisskopf (195D has deduced a formula for the radiative transition probability of an excited nucleus assuming' that the radiation results from the de-excitation of only one particle i n the f i e l d of the others. Wilkinson (1953) has compiled an impressive l i s t of agreements between predicted and experimental values of this quantity i n reactions of the type A(b,»)C. However, i n view of some recent suggestions that the elementary particles may loose their identity within the nucleus, i t i s well to consider what properties could be a test of the independent particle model. One such process i s that of direct radiative capture of a proton by the especially stable nuclear configurations i n O1^ and C 1 2. The radiation - 3 -can "be explained by considering the proton captured i n an atom-like orbit around a passive core vaith de-excitation of the proton alone. In other nuclei the capture radiation process appears to be more complex, but a study of such transitions should indicate at least the limitations of the model. The f i r s t part of this thesis describes the investigation of the capture radiation from the reaction B 1 0(p,*)C 1 1, using a sodium iodide s c i n t i l l a t i o n counter. The work of the second part began as a search for capture radiation from the reaction B 1 0(d,OC 1 2. Although the search was unsuccessful, the competing reactions Bl°(d,p^)B-1:-L and B 1 0(d,ny)G 1 1 provide another test of the shell model. The residual nuclei and are mirror nuclei, differing only i n the exchange of one proton and one neutron. The simultaneous measurement of the energies and relative intensities of the radiations from these nuclei should deter-mine i n what respects the energy levels are affected by the exchange of particles. Accordingly, the gamma ray spectra of these reactions were determined. If the nuclear forces are largely charge independent, the energy level structure of mirror nuclei would be very similar. The spins and parities of corresponding levels should be identical, their energies differing only slightly through the effects of Coulomb forces. - if -Experimentally, the problem of detecting the radia-tions with s c i n t i l l a t i o n counters i s made d i f f i c u l t by the large number of transitions possible and the interaction i n the crystal of the neutrons produced with the gamma rays from the reaction B 1 0(d,nif)C 1- L, A number of prompt gamma rays and delayed beta particles are produced by the absorption of neutrons by iodine, so that the pulse height spectrum from the photomultiplier i s extremely complex. With the use of a three crystal pair spectrometer of improved design, i t was possible to resolve the gamma rays associated with the reactions mentioned from the background radiations. It i s with the construction and operation of the spectrometer that the second part of this thesis i s concerned. F i e . 1. Energy leve l s i n the nucleus C l l . - 5 -CHAPTER II DETECTION OF THE RADIATIONS FROM THE PROTON BOMBARDMENT OF BORON TEN 1. HISTORY AND EXPECTATIONS It i s only natural that the nuclei whose structure has been most studied are those formed by the bombardment of the more abundant isotopes. Only the most p r o l i f i c or highly exoergic reactions involving the bombardment of the relatively less abundant isotopes have been observed; the weak reactions have had to await the production of isotopically pure targets. For this reason the bombardment of B^ 0 with protons can s t i l l yield important information. Previously masked by the high energy radiations from the reaction BJ"L(p,^)C , the reaction B 1 0(p,2T)C 1 1 was not observed u n t i l Walker (1950), using a magnetic pair spectrometer, resolved a 9.1+7 Mev. gamma ray from the bombardment of a thick enriched B^ 0 target with 1,16 Mev, protons. The excitation function of this radiation was subsequently studied by two groups of investigators with conflicting results. Krone and Seagondollar (1953) reported resonances at Ep = 0,78, 0,95j and 1.33 Mev,, while Day and Huus (195^) reported only one at E p r 1.2 Mev, and possibly another at E p = 2,h Mev. Other information pertinent to this investigation i s summarized i n F i g . 1 i n the p i c t o r i a l representation of Ajzenberg and Lauritsen (1955). The horizontal lines representing energy levels are marked with their energies i n Mev.,.referred to the C±Jm ground state as zero. Particle transitions are indicated by sloping lines, radiative transi-tions by vertical lines. The energy differences between the ground state and those of the other nucleon combinations are calculated from the isotopic mass data compiled by L i et a l . (195D• Excitation functions of the various reaction products are indicated beside the corresponding, levels i n C^. . Two groups of alpha particles from the reaction 10 7 B (p,«)Be' were resolved by Brown et a l . (1951) corresponding to the formation of Be7 i n i t s ground and f i r s t excited states. The ground state alpha particles, marked or0 i n Fig. 1, show resonances at Ep = 1.1 Mev. ( c = 11 mb/ster. at 138° to the beam) and Ep = 1.5 Mev., while the second group, « (, has only one resonance at Ep = 1.5 Mev. with the same peak cross section. The k30 kev. radiation from the de-excitation of the f i r s t excited state of Be7 has, of course, the same yield function as the «i group, with a peak cross section of 0.21 bn., i.e. 16 mb/ster. (Day and Huus 195*0. The present study can be expected to resolve the con-f l i c t in evidence for the excitation function of the reaction B" 1" 0(p,y)C 1 1. A knowledge of the angular distribution of the 9 Mev. radiation would also be useful i n order to calculate the total cross section of the reaction, and to deduce the spin and parity of the energy level of C^- responsible for the resonance. A measurement of the energy of the capture gamma ray would allow the Q of the reaction to be calculated and hence the mass of the beta -unstable nucleus C . Finally, radiative transitions to levels other than the ground state should be present, because they appear when i s excited through the reaction B^CdjntfC 1 1, and also i n the mirror nucleus B 1 1 when i t i s excited through the reactions B^ 0(d,p*)B i : L and Li 7(«, OB 1 1. 2. APPARATUS ' ( a) The target The yield of gamma rays from the reaction B 1 0(p,v)c i : L i s rather small as i n a l l cases of radiation i n competition with particle emission, and special precautions were taken to eliminate background radiations. A very pure target of B^ 0 was used, and kept free of contamination i n the vacuum system. Fluorine i s an especially serious contamination because the p r o l i f i c 6 Mev. radiation from the reaction Fl9(p,cty)0 1 6 from one part of F^9 i n 1C-5 i n the target could mask the lower energy cascade transitions expected to be present. The target backing material was carefully chosen to reduce possible background radiations. Metals of large atomic number w i l l not yield high energy radiations because the proton penetrability of their nuclei i s small, but some are easily excited by the electric f i e l d of passing particles; the so-called "Wolfcall" radiation studied, for example, by McClelland et a l . (1955). Though of energies far less than that of the gamma rays to be studied, these radiations can be quite p r o l i f i c and produce "pile-up" pulses i n the detector which appear, as background i n higher energy regions. - 8 -Most of the base metals contain impurities such as sodium, fluorine, and calcium, a l l p r o l i f i c sources of radia-tion under bombardment. Even i f pure, they tarnish easily and are hard to keep clean. The noble metals, however, are by their inertness not easily contaminated and are obtainable i n very pure form. Gold and platinum contain a minimum amount of impurity and are not easily excited e l e c t r i c a l l y . Tungsten i s even better i n this respect for protons of energy below 1 Mev., but much worse at higher bombarding energies. Targets of magnetically separated B^, hOO /»g/cm2 and 250 jtg/cm2 thick deposited on gold and platinum, were kindly supplied by the Isotopes Division, A.E.R.E., Harwell. They were mounted for bombardment i n a target chamber described i n detail elsewhere (Alexander 1955). The wall of the chamber was brass, 1/32 inch thick, and the beam entrance tube was lined with lead to absorb radiations from the bombardment of the stops defining the beam. The stops were gold, electrolytically etched i n a potassium cyanide solution to remove contaminants, but carbon and fluorine found their way onto them from the vacuum system after a few hours. The target chamber could be isolated from the rest of the vacuum system by closing a large valve i n front of the entrance tube; i t was closed whenever the target was not being bombarded, (b) The S c i n t i l l a t i o n Counter The high efficiency and energy resolution of the sodium iodide s c i n t i l l a t i o n counter make i t the best gamma ray - 9 -spectrometer available for the study of weak radiations. The properties of such detectors have been discussed at length elsewhere (Mclntyre and Hofstadter 1950, Griffiths 1953 > Azuma 1953)» "but a few remarks pertinent to the detection of high energy radiations should be noted. Because of the fast electrons and secondary quanta produced in the absorption of gamma ray energy i n sodium iodide may escape the crystal with any amount of energy up to the f u l l energy of the incident gamma ray, the height of the voltage pulse from the counter, proportional to the energy absorbed, can have any value up to a maximum corresponding to complete absorption. The secondary quanta are the scattered gamma rays from the Compton effect and Bremsstrahlung quanta from the fast electrons produced by a l l interactions of gamma rays with Nal. According to Heitler (195^} P. 2*1-2) the rate of energy loss of a 10 Mev. electron i n sodium iodide is about equally by Bremsstrahlung and by ionizing collisions, with the ratio rapidly increasing i n favour of Bremsstrahlung with electron energy. Eighty percent of this energy loss i s i n the form of quanta of over 1 Mev. energy, which can easily escape the crystal. Clearly, the larger the crystal used, the more often w i l l the total gamma ray energy be absorbed. However, the largest crystal available for the experiment was a cylinder 2 inches long by 1^ inches diameter. Because these dimensions - 10 -set the ultimate s c i n t i l l a t i o n intensity distribution, the problem then was to reproduce this distribution as f a i t h f u l l y as possible i n the voltage pulse height distribution from the photomultiplier. After some experimentation i t was found that improved techniques i n mounting the crystal on the photomultiplier would give better energy resolution. As a criterion, the resolution of the pulse height distribution resulting from the detection of the 1.28 Mev. gamma ray from Na 2 2 was adopted as a test of the crystal mounting technique. Resolution was defined as the ratio of the f u l l width at half maximum height of the pulse group corresponding to absorption of the f u l l gamma ray energy to the voltage of the pulses producing the maximum height. Details of the mounting procedure f i n a l l y adopted are presented i n the appendix. The use of newly available photomultipliers resulted in further improvement i n resolution. Dumont 6292 and RCA 63*+2 photomultipliers have an increased photocathode efficiency (number of electrons released by the photocathode per Mev. energy released i n the phosphor). Using the formula of Roberts (1953) for energy resolution of a s c i n t i l l a t i o n counter and the experimentally measured energy resolutions i n Table I, i t was deduced that 1500 electrons were released by the photocathode of an RCA photomultiplier per Mev. energy dissipated i n the phosphor. Griffiths (1953) showed that for the older EMI 6262 photomultiplier, only 1100 electrons per Mev. were released. 2 0 0 0 V O L T S U P P L Y o 3 0 0 V . -PULSE ^H- P U L S E RCA 6 3 4 2 P H O T O M U L T I PLIER H E A D A M P L I F I E R F i g . 2.  - 11 -Table I Resolution of Nal s c i n t i l l a t i o n counter, with crystal 2" x 1-|" diam., RCA 63*f2 photomultiplier Source Gamma Energy Resolution E u 1 ^ 0.085 Mev. 20-25$ Na 2 2 0.51 Mev. 8.3$ Na 2 2 1.28 Mev. 7.8$ ThC" . 2.62 Mev. 6.6$ (c) Electronics The counter head amplifier and high voltage supply are shown i n Fig. 2. Power for the cathode followers was provided by a Lambda Inc. model #28 power pack. The RCA 63*+2 photo-multiplier was run with 975 volts across the bleeder resistor chain, which kept the f i r s t dynode potential at 230 volts above that of the photocathode. The focus electrode was kept at the same voltage, as recommended by RCA for best resolution. Curiously, maximum gain was obtained with the focus potential at about 80$ of this value, but not best resolution. Both negative and positive pulses could be obtained from the collector and last dynode respectively. The collector pulse was of about 6 psec, duration, the dynode pulse about 3 jusec. The negative pulse was amplified by a Northern El e c t r i c #lW+ linear emplifier capable of producing 50 volt linear pulses. The amplified pulse was fed into a biased amplifier which amplified only that part of the pulse greater than a - 12 -variable bias level, so that when the output was driven into a Marconi type #115-935 thirty channel pulse height analyser or "kicksorter", the effective bias level of the f i r s t channel could be set to any value. The operation of the biased amplifier i s explained i n Chapter III, the circ u i t diagram shown i n Fig. 13. The positive dynode pulse was used to monitor the total counting rate i n the counter by feeding i t into an Atomic Instruments, Inc. #20 -^0 linear amplifier, the discriminator output of which was driven into a scaler. 3. EXPERIMENTAL PROCEDURE The targets were bombarded with proton beams from the U.B.C. Van de Graff generator, with magnetically resolved energies from 0.5 to 2.0 Mev. The beam flux was measured with a current integrator (Edwards 195D > calibrated and found not to vary more than 3 percent for any current input. The proton energies were determined by measuring the accelerating voltage with a generating voltmeter calibrated with the well known resonances of the F ^ t p ^ i O O 1 ^ reaction. Gamma ray energies were determined by calibrating the pulse height response of the s c i n t i l l a t i o n counter with gamma rays of known energy. These included the 0.511 and 1.28 Mev. gamma rays from Na 2 2, the 2.62 Mev. gamma ray from ThC", and the 6.13 Mev. gamma ray from the reaction F19(p,otV)01^ at Ep = 935 kev. An occasional further calibration was made with the 9.17 Mev. gamma ray from the reaction C 1 3(p , O N l l f at - 13 -E p = 1.76 Mev. (Woodbury et a l 1952), but since the C"13 target was very thin, i t was not often used. The linearity of the energy-pulse height calibration was satisfactory i f the counter was run at 975 volts, but became non-linear at higher voltages. It is believed that this limitation was a property of the * particular photomultiplier used, as others gave a linear response up to 1500 volts H.T. To reduce background, the counter was encased i n a lead block cast to enclose i t , and offering six inches of lead between the crystal and the magnet vacuum box. Furthermore, the target area counter was shadowed by four inches of lead two feet above i t . With these precautions, the background from cosmic rays of over 10 Mev. energy was 18 per minute. In the measurement of the angular distribution of the 9 Mev. radiation, the counter and i t s shielding were rotated together. A second s c i n t i l l a t i o n counter, a 1 inch long by diameter sodium iodide crystal mounted on a Dumont 6292 photomultiplier was set at 135° to the beam direction, and also encased i n lead. The pulses from this counter were amplified by a second Atomic Instrument Inc. #20*f - C linear amplifier with the discriminator set to trigger a scaler on pulses representing over 7 Mev. energy release i n the phosphor, and the amplifier output fed into a scaler with discrimination level set to bias out pulses representing less than 10 Mev. energy release. In this way the major cosmic ray background could be subtracted with no increase i n s t a t i s t i c a l uncertainty. CHANNEL NUMBER - Ih -Preliminary runs were spoiled by large gain d r i f t s i n the photomultiplier caused by excessively fast counting rates for the ^30 kev. radiation from the reaction B 1 0(p , « O B e 7 . The gain increased rapidly under the gamma ray flux, but • returned to normal i n a few hours. Such behavior has been recently reported by Caldwell and Turner (1951*). Probably large currents i n the photomultiplier either heated the dynodes or caused ion bombardment of the photocathode and dynodes from the residual gas i n the envelope. The latter explanation i s supported by the fact.that one R.C.A. 63H-2 showed nnjch greater st a b i l i t y than most of the other photomultipliers and also could withstand 2,200 volts across i t without breaking down. Unfortunately, this tube was not available for the experiment, so that the counting rate of pulses representing over O.U- Mev. energy release i n the phosphor had to be kept below 3j000 per second to maintain gain s t a b i l i t y . h. RESULTS (a) The 9 Mev. Radiation Fig. 3 shows a typical pulse height distribution from the counter due to the radiation from the reaction B 1 0(p,*')C 1 1 at 1.0 Mev. proton energy. The presence of 9»6h Mev. radiation is obvious; the rise at 7.91 Mev. is not attributed to a separate gamma ray, but rather to wall effect. The cross section for production of soft Bremmstrahlung quanta from fast pair electrons i n the crystal does not rise as sharply as the absorption coefficients of Nal as the energy decreases, so that - 15 -a maximum w i l l occur between Bremsstrahlung production and escape. This effect has been observed i n the three crystal spectrometer pulse height distribution from the pure 9 Mev. radiation from the reaction Fl9(p,ft*)0 i 6 (Fig. 1>+). Such spectra were obtained for proton energies from 0.5 to 1.6 Mev. i n 100 kev. steps, with the counter at 90° to the beam and i t s face 10 cm. from the target. The gamma ray energy varied with proton energy i n the manner expected for a simple capture process, i.e. E = Q + 10/11 E p . a v . where Ep_ a v was taken to be the incident proton energy less one-half that lost i n the target, since the cross section for the reaction varies quite slowly with proton energy. The proton energy loss i n the target was assumed to be the same as that for an equal mass per cm2 of beryllium, and the measurements by Masden (1953) were used to calculate i t . This procedure seems to be valid because the theoretically predicted values of stopping power for the two elements differ by less than 2$ (Aron et a l . 19^9). In the \QO jag/cm2 thick target placed at ^ +5° to the beam, a one Mev. proton then lost 126 kev..energy. The mean Q value found from ten gamma ray energy measure-ments was 8.81 Mev. with a standard error 0.02 Mev. This value is 0.11 Mev. greater than that calculated from the masses of B^, the neutron-hydrogen atom mass difference, and the 3.015 Mev. threshold of the reaction B 1 1(p,n)C 1 1 (Li et a l . 195D. 4 PROTON ENERGY (MEV.) Fie, M-. Excitation function of the 9 Mev. radiation from Bl°(p,/)C1:L. - 16 -Then mass of B 1 1 = 10.01611^ amu. plus mass of H 1 =- 1.0081^ -2 amu. minus Q = 8.81 = .009^61 amu. 931.152 gives mass C 1 1 = ll.OlM-795 amu. as compared to llt011+925 amu. adopted by L i et a l . It may he noted that there is also some discrepancy between this mass value and that found from the maximum beta ray energy of the Cll(a+) B 11 spectrum (.968 Mev.), which i s 11.01^927 amu. Fig. h shows the excitation function of the 9 Mev. radi-ation. Here proton energy is corrected as before for the target thickness, and the total cross section is corrected for an angular distribution of the form 1+0 .5 cos 2©. A background of roughly 10$ i s subtracted, mostly from cosmic rays because the runs were of long duration, but a small beam dependent part was found from radiations over 10 Mev. energy which, i s due entirely to the reaction B ^ l p ^ ) ^ 2 , showed that there was less than one part i n 300 of B 1 1 i n the target. Since the gamma ray energy varied with proton energy, i t was necessary to choose a new IVbias" point on the gamma ray energy scale for each proton energy above which to count pulses. This was determined by measuring back from f u l l gamma ray energy an interval of 2.*+ Mev. on the plotted spectra i n order to include as many pulses as possible without counting i n the background from lower energy radiations. It also happens that measuring back a fixed amount is practically equivalent to fixing the minimum angle that a Compton scattered photon can - 17 -be deflected and s t i l l give a countable pulse. The ratio of Compton pulses above and below this energy i s then constant, and so the "bias efficiency" should not change more than 2>% over a one Mev. photon energy range. The excitation function shows a maximum at Ep = 1135 i 15 kev. The f i n i t e yield above 1.5 Mev. was assumed to be nonresonant and was continued to zero at 0.*+ Mev. to estimate the width of the resonance. The width at half maximum yield then was 5*+0 ± hO kev. The angular distribution of the 9 Mev. radiation at the resonance energy i s shown i n Fi g . 5. With the rotating counter crystal face 20 cm. from the target, the solid angle of the detector was that of a 6° cone, and solid angle corrections were unnecessary within the accuracy of the measurement. Different symbols i n Fig. 5 indicate which runs were bracketted by runs at 90° and reBormalized. The large block of lead used for shielding prevented measurements at angles to the beam of greater than 127°. In the time allotted for the experiment only a rough measurement was possible; to obtain 1500 counts in the rotating counter and 1000 i n the monitor took about an hour and gave h% s t a t i s t i c a l uncertainty. The ratio of the yield at 0° to that at 90° was measured several times and found to be 1.5 * 0.05. A least squares f i t of a function to the data of Fig. 5 would be misleading because the uncertainties of the points are too great. A small asymmetry around 90° cannot be definitely ruled out, but i f a distribution of the form 1 +• Aicos'0 + A^cos2© - 18 -i s assumed, A l < 0.1. A preliminary measurement also indicated that the asymmetrical term i s even less. It would he best to remeasure the angular distribution of the radiation at various bombarding energies, and to arrange to measure the intensities at angles greater'than 127° in order to detect any interference between formation of the level of C 1 1 producing the resonance at 1135 kev. and any adjacent levels. The continuous rise of the yield above the resonance suggests that interference phenomena are possible. The absolute cross section 6~ for the 9 Mev. radiation i s given by: 6" = n y npn tit .4 wKere »j= total gamma ray interactions i n the counter 1%= total number of protons »t = number of target atoms/cm2 -fl- = solid angle of counter € = efficiency of counter. Since a l l gamma rays from the target do not pass through the same length of crystal, a numerical integration of the following integral must be performed to calculate the product Jlf for the crystal-geometry: - 19 -where = absorption coefficient for the radiation = .138 cm-1 for 10 Mev. photons Jtfr) = the graphically measured gamma path length at the angle 9 to the crystal target axis Jito = the solid angle between 0 and 0 + d9 = sinOd© spheres. For a distance from the crystal face to the target of 10.2 cm. s 0.00^28, in units of spheres. The total counts represent the total number of inter-actions regardless of secondary processes and so the efficiency can be deduced from the cross sections for these interactions (Davisson and Evans, RMP, 2|±, 79, 19.52). Since there i s no pure 9 Mev. radiation available to ascertain the shape of the spectrum, i t was necessary to assume a form of " t a i l " to the three peaks to deduce the ratio of counts above the "bias level" to the total number i n the spectrum. The validity of the general procedure has been checked experimentally by calibrating the counter efficiency with the known yield of 6 Mev. radiation from the bombardment of a thick CaF2 target with protons. Both methods agreed within 3% that 30$ of the 6 Mev. gamma ray flux through the crystal released over h Mev. energy (Alexander 1955). It was assumed that the ratio of the height of the t a i l to the height of the peak was the same for the 9 Mev. as for the 6 Mev. gamma ray spectrum, so that 0.*f0 of the pulses are above the chosen bias level. An upper limit to this figure can be estimated by continuing the spectrum smoothly to zero at zero - 20 -pulse height; i t i s 0.55* A lower li m i t , obtained by assuming that the minimum point on the experimentally obtained spectrum continues at this height to zero i s 0,3^. It is then improbable that the procedure should introduce more than a 20$ error. At the maximum yield there were 1.10 counts per micro-coulomb above the "bias level". Thus correcting the cross section for the angular distribution 1 + 0.5 cos^O, <s « (3.5 1 1.0)10-30cm2. The rather large error i s an estimate of the uncertainties i n both counter efficiency and target thick-ness. (b) Intermediate Energy Radiation Fig. 6 shows the gamma ray spectrum from 1.5 to 10 Mev. with background subtracted. Gamma rays are clearly present, but are not well enough resolved to be assigned accurate energies. Three gamma rays of energies 6.5, ^.8 and ^ -.3 Mev. appear resolved and might be attributed to cascade' transitions through the levels i n C u at 6.35, ^.77 and 1+.23 Mev. The prominence at 2.5 Mev. would indicate radiations of about this energy as well. To resolve the decay scheme completely, however, would require counting the gamma rays i n coincidence. (c) Low Energy Radiation The excitation function of the ^30 kev. radiation from the reaction B 1 0(p ,aOBe7 reached a maximum at 1575 kev. bom-barding energy, or 1525 kev. energy, corrected for target thickness. These figures agree almost exactly with those of CHANNEL NUMBER _ n  Fie. 7. Lower energy radiations from B 1 U -»-p. Explanation in text. - 21 -Day and Huns (195*0 > indicating that the target thickness assumed is probably correct. Fig. 6 shows the spectrum of radiation with energies from 0.6 to 1.5 Mev. Curve A shows a run at E p = 1.2 Mev. using the *f00 j*g/cm2 thick B^ -0 target on gold. Curve B shows a run at E p = 1.1 Mev. using the 250 jxg/cm2 thick target on platinum, while Curve C is the background from platinum. Curve D is the energy calibration: the 1.28 Mev. gamma ray from Na 2 2. At Ep =1.2 Mev. a 720 kev. gamma ray appeared, presumably from the de-excitation of the f i r s t excited state of B 1 0 i n the reaction BlO(p,p'>')B10 (Day and Huus 195*0. It was not visible at E p = 1.1 Mev., perhaps because the back-ground was larger from the platinum ;target. At both energies and with both backings radiation of energy 850 kev. was present. Since i t s intensity i s roughly the same as that of the 9 Mev. radiation, i t may be tentatively assigned to a cascade transition i n the reaction B 1 0(p,Jr)C i : L. 5. Discussion (a) Widths of the Excited State of C 1 1. The preceding results show that the bombardment of B 1 0 with protons of around 1< Mev. energy leads to a resonant yield of 9 Mev. radiation which has been attributed to the reaction B 1 0(p,y)C 1 : L. Since the resonance energy and yield curve shape are the same as that measured by Brown et a l . for the reaction B 1 0(p,<0Be 7, (Brown 1950), i t i s f a i r l y certain that the same excited state of C 1 1 i s involved i n both reactions. - 22 -With this information i t i s possible to calculate the transition probability for the radiation. Assuming that the resonance yield i s governed by the Breit-Wigner dispersion formula, the'cross sections (Trf and 6Y for the two reactions at the resonance energy are: wkere ct> = 2J»1 , a s t a t i s t i c a l factor (2s+l)(2i+l) J = spin of the excited state s r spin of the incoming proton i = spin of the target nucleus A = reduced wave length of the incoming proton rj, = partial width for reemission of the proton Tat = partial width for emission of an alpha particle IV = partial width for emission of a gamma ray r = total width of the reaction - Ton- r,. + Ty = measured half width of the resonance curve. The partial width J~$ i s related to. the transition probability T by: P Y ( i n ev.) = 6.55 x l ( T l 6 T ( i n sec" 1) It can be seen from the formulae that Prs^^r/^ec so and hence 1"^  must be determined. Assuming that the ground state alpha particles are isotropic, we can use for o«. the value found for the yield of alphas at 138° by Brown et a l , i.e. 16 x 10~27 cm 2/steradian. Then - 23 -Since to««l, the radical may be expanded and It therefore follows that Tot ^ T - r f ~ 500 kev. and r ^.ftclQ-30 x ^ k e v > „ 9 e v > * M-«l6xl0-^/ It is interesting to compare this value with that pre-dicted by the independent particle model of the nucleus (Weisskopf 195D for 9:.8 Mev. radiation: I"V = V70 ev. for electric dipole radiation = 20 . ev. for magnetic dipole radiation = 0.23 ev. for electric quadrupole radiation = 0.001 ev. for magnetic quadrupole radiation. As was mentioned i n the introduction (Chapter I ) , Weisskopfs formula cannot be relied upon to distinguish between possible multipolarities of gamma radiation, but w i l l provide at best an upper limit on the radiative width of the transition. It. is evident that magnetic dipole radiation would i n this case be quite reasonably expected. (b) Spin of the Excited State of C 1 1 •Speculation about the spin and parity of the excited state of C 1 1 responsible for the resonance i n the reaction BlO(pi* )C1JL at Ep = 1135 kev. i s tempting but on the rather Incomplete evidence obtained may well be entirely wrong. It is possible, however, to state that the spin of the excited state i s not 1/2, nor do s-wave = 0) protons i n i t i a t e the reaction. More positive statements are not possible without some sweeping assumptions. - 2h -The f i r s t d i f f i c u l t y is that the spin of the ground state of C 1! has not been measured. However, i t is predicted to he 3/2 (-) by the shell model theory, an assignment supported by the fact that the mirror nucleus has a measured spin 3/2. An analysis of the angular distribution of the ground state group of neutrons from the reaction Bl°(d,n)cH by stripping theory (Paris and Endt 195*0 indicates that i n the reaction the ground state of C 1! is formed by £-g - 1 protons, and since the spin of B 1 0 is 3(+), and of the proton 1/2 (+), the parity of C 1 1 should be odd. An analysis was therefore carried out on the assumption that the ground state spin of C 1 1 i s 3/2 (-) to determine what spin assignments to the state excited in the reaction B 1 0(p,y)C 1 : i- would be consistent with the experimental data. The compound state was assumed described by only one value of spin and parity. The tables of Sharp et a l . (195*0 were used i n the analysis. When the proton is absorbed by the B 1 0 nucleus, the spin of the proton, the spin of the B 1^ nucleus, and the orbital angular momentum of the proton add vectorially to produce the spin J*of the compound state. It is customary to consider the sum of the f i r s t two spins as defining the "channel spin" s* and the "channel parity". In this case s can have two values, s i = 5/2 and S2 = 7/2. The orbital angular momentum £^ of the proton then is added vectorially to ~£ to give J*, and incidentally defines the parity of the compound state. In most cases J may - 25 -result from the addition of either s^ or s^> to ^/p, and the two possible modes of formation give rise to two incoherent contri-butions to the yield, i.e. ¥(0) = W si(e) + tW s 2(9) where t is the "mixing parameter", determined by the sensi-t i v i t y of nuclear forces to the angular momenta of the particles. Unless specific assumptions about the nuclear forces are made, t can have any positive value. The terms W S T_ and WS2 are each i n turn made up of the coherent contributions to the yield from the various possible multipole radiations which can de-excite the compound state... If the residual nucleus has spin s*1 the gamma ray can take away any-angular momentum J# such that ^ *"s' = 7. For example, i f electric dipole ( ^ B 1) and magnetic quadrupole 2) rad-iation is possible, Wsi(e) <=*- W E I ( O ) + aWM2(©) + D W I N T - 1 , 2 W E4. esc % / r ocJZ ZC^pJ-/pJ,sk)Zi(4'J-4j,sIk)Pk(cflsfi) k WlHT-l,2c*5j Z(/pj/ pJ,sk)Z 1(4.ij4^JjS'k)P k(cos9) k where Z and Z\ are the coefficients tabulated by Sharp et a l . (195*+) and PkCcos©) is the usual Legendre polynomial. The weights a and b are determined by purely nuclear" properties; i f they are l e f t arbitrary the distribution is indeterminate. For the sake of simplicity i t w i l l be assumed that the radiation i s of only one multipolarity. If the sum for W(9) is reduced to the form 1 + Acos 2© •*• Bcos1*©, i t i s simpler to - 26 -compare to the experimental data. Table II shows these coefficients for the various values of J which can be formed i f /v = 1 or 2. TABLE II Coefficients i n the pure multipole radiation angular distributions possible for various assumed values of spin and parity of the excited state of e l l , Ws(©)<~ 1+ Acos2© * Boos4-© Spin J of C 1 1 Parity ¥ of C 1 1 V p Radiation type Dist. for 8 l = 5/2 Dist. for s 2 = 7/2 3/2 + 2 E l M2 E3 Bl AP -0.37 0 -O.U-9 0 0 0 0.18 0 0.28 0 0 0 3/2 — 1 Ml E2 0.125 0 0 0 / X X X X 5/2 -+• 2 E l M2 0.23 -l.h 0 1.5 O A l 0.27 0 -0.53 5/2 — 1 Ml E2 0.57 -0.23 0 0 -O.lh 0.08 0 0 7/2 2 M2 E3 -1.07 0.06 1.37 -0.08 0.59 -0.32 -1.09 - 0 . 0 5 7/2 — 1 E2 M3 0.65 0 0 -0.51 -0.58 0 . 0 9/2 -f 2 E3 l.lh . -0.33 0.^ +9 -0.52 9/2 — 1- Mg x x 0-.89 0 x = state cannot be formed with this value of s The f i n a l distribution W(0) ^ W S l(0) +• t¥ S 2(©) w i l l be of the form W(©) «*. 1 + Ccos 2© + Dcos^©, where min(A1,A2) ^ C max(A1,A2) with a similar inequality for D. Since experimentally C = 0 . 5 5 - 27 -J cannot be 3/2 or 9/2. J =7/2 can also be discarded. For J =. 5/2 (+), E l ra d i a t i o n , C ^ ©Al, which i s just outside the experimental uncertainty l i m i t s . . Both J s 5/2 (-), Ml radi a t i o n , and J = 7/2 (-), E2 or M3 rad i a t i o n , could produce the experimentally measured angular d i s t r i b u t i o n . The r a d i a t -ive width of the t r a n s i t i o n , however, indicates that J = 5/2(-), Ml radiation, i s the most l i k e l y p o s s i b i l i t y . I t must be born i n mind that such an assignment cannot be considered d e f i n i t e since the p e s s i b i l i t y of interference between the formation of states of opposite p a r i t y i n the com-pound nucleus has yet to be explored. Considerably more study of both the gamma and alpha radiations from the proton bombard-ment of B 1 0 should be undertaken. CHANNEL NUMBER - 28 -CHAPTER III THE THREE CRYSTAL PAIR SPECTROMETER 1. Principle of the Spectrometer The preceding chapter has illustrated the d i f f i c u l t i e s involved i n the investigation of reactions i n which many-different energy gamma rays are produced. In the pulse height distribution from a sodium iodide s c i n t i l l a t i o n counter due to gamma rays of less than one Mev. difference i n energy, the overlapping pair peak triplets cannot be distinguished with any accuracy, and gamma rays of small intensity can be obscured by the " t a i l " of a more intense, higher energy gamma ray. If neutrons are also produced i n the reaction, they w i l l create a background of pulses which w i l l confuse the spectrum further. When a neutron i s absorbed by 1127 prompt gamma rays of energies from two to four Mev. and delayed beta particles of energies around two Mev. are emitted, and when absorbed i n the crystal can produce a pulse equivalent to that of a gamma ray of as much as 15 Mev. energy. As an i l l u s t r a t i o n of the d i f f i c u l t y , a single crystal sodium iodide s c i n t i l l a t i o n counter pulse height distribution resulting from the deuteron bombardment of B^ 0 i s shown i n Fig. 8. The previously observed groups of gamma rays from the reactions B 1 0(d,hOC 1 1 and B 1 0(d,pJf)B 1 1, superposed on a large background, are evident, but the distinguishing peaks are unresolved. A magnetic pair spectrometer could be used to CATHODE FOLLOWER N4Cn) 6262 D I F F E R E N T I A L D I S C R I M I N A T O R COLLECTOR V 6262 CATHODE FOLLOWER 6342 DYNODE C A T H O D E F O L L O W E R D I F F E R E N T I A L D I S C R I M I N A T O R C A T H O D E F O L L O W E R D E L A Y D E L A Y N O R T H E R N E L E C T R I C A M P L I F I E R A M P L I F I E R A N D L I M I T E R G A T E A N D B I A S E D A M P L I F I E R I T R I P L E C O I N C I D E N C E M I X E R G A T E P U L S E G E N E R A T O R K I C K S O R T E R C E N T R E C R Y S T A L 2 * 4 x 5 cm. S I D E C R Y S T A L S l"x I T " dlam. Block diagram of the three crystal pair spectrometer. - 29 -resolve the radiations, but with a very gre„at loss i n detection efficiency. If i t were possible to record only those pulses from a sodium iodide s c i n t i l l a t i o n counter corresponding to gamma ray energy release through the pair effect, a much more efficient pair spectrometer would result. Previous work done i n this laboratory has shown that such a technique i s indeed possible. The escape of annihilation quanta from a s c i n t i l l a t i o n crystal can be used to signal the fact that a pair event has occurred. If the signal i s used to gate the pulse from the photomultiplier, a spectrum of only one peak per gamma ray, the pair peak, w i l l be displayed on the kicksorter. Bair and Maienschein (195D, Johansson (1952), and Griffiths (1952), have constructed such instruments, the latter in this laboratory. A block diagram of a three crystal pair spectrometer, modified and used by the author, is shown-in F i g . 9« Two sodium iodide s c i n t i l l a t i o n counters, the "side channels", were placed one on either side of the "centre crystal", a block of sodium iodide of dimensions 5 cm. by h cm. by 2 cm. mounted on an RCA 63^ +2 photomultiplier and composing the "centre channel". The side crystals were cylinders of Nal 1 inch long by l i | inch diameter. The centre crystal was placed with i t s least dimen-sion separating the side crystals. When a pair event occurs i n the centre crystal, the colinear annihilation quanta produced - 30 -have a good chance of escaping and being captured simultaneously i n the side crystals. The pulses produced i n each of the counters are put into a tr i p l e coincidence c i r c u i t and the coincidence pulse then provides the signal that a pair event has occurred i n the centre crystal with the escape of the annihilation radiation. It triggers a gate which allows the centre channel pulse to enter the kicksorter. Originally a l l pulses from the side channels were put into the coincidence c i r c u i t , and shielding of the crystals, a short (0.1 usee.) resolving time of the coincidence c i r c u i t , and the colinearity of the annihilation quanta were relied upon to reject spurious ' events• However, i f a large flux of radiation is present, real coincidences not due to pair events i n the centre crystal can occur. If a gamma ray i s scattered i n the centre crystal through the "double Compton" effect (Mandl and Skyrme 1952), the two quanta produced can enter the side crystals and cause a coincidence to occur. Single Compton quanta can be scattered from one side crystal to the other with the same effect. In the presence of neutrons, shielding is rather ineffective, and many neutron capture gamma rays are produced, often i n cascade. If accidental coincidences open the gate and a pulse i s also produced not quite, simultaneously i n the centre channel, the centre channel pulse can be distorted by the gate and placed in any part of the kicksorter pulse height spectrum. - 31 -In order to produce a true pair spectrum, i t was decided to use differe n t i a l energy discrimination i n the side channels to allow coincidences to occur only between pulses of a size corresponding to within 50 kev. of 0.51 Mev. energy loss i n the side crystals. This system has been tried with success by West and Mann (195^)• It means rejecting about 90$ of the true pair events, since only 27$ of the 0.51 Mev. gamma rays captured i n the side channel crystals i n i t i a t e " f u l l energy" pulses. A further rejection of accidental coincidences, this time with no loss of efficiency, was achieved by demanding that the pulses from a l l three counters be i n coincidence before the gate was opened. The loss of efficiency through side channel energy discrimination i s amply compensated for by the increased resolu-tion and background rejection. Most neutron capture gamma rays are about 8 Mev. i n energy, and so w i l l contribute few pulses to such a small region around 0.5 Mev. Double Compton quanta are mostly of unequal energy (Heitler 195^ > P. 22*f); twice scattered quanta w i l l always be degraded below the energy selection l i m i t . Pulses from the gamma rays produced by neutron capture i n the centre crystal cannot be entirely rejected, but only if.the gamma ray interacts i n the centre crystal by the pair effect is there much probability that a pulse w i l l be recorded. The pulse i n the side channels from the capture of gamma rays cascading with the one captured i n the centre crystal w i l l rarely be of a height that w i l l be accepted by the side channel - 32 -discriminators. Since the pair production cross section of Nal is f a i r l y small for the neutron capture gamma rays individually, they w i l l he discriminated against. The use of tr i p l e coincidence mixing i n place of double reduces the rate of purely accidental gate openings from 2BjE2'c to ^ N T J ^ ^ T 2 , where NT_, N2, N3 are the counting rates i n the channels and T i s the resolving time of the coincidence c i r c u i t . Since the permissible counting rates must be limited i n order to maintain gain s t a b i l i t y , as was discussed i n the last chapter, an extremely fast coincidence c i r c u i t was unnecessary; one microsecond resolving time was sufficient to reject almost a l l accidental coincidences. 2. Details of the Spectrometer In the block diagram of the three crystal spectrometer in Fig. 9» the "centre channel" is indicated by the heavy line connections between the blocks. The centre counter collector pulse was driven by a cathode follower in the head amplifier into a Northern Electric #lW+ linear amplifier, then into the gated biased amplifier from which, i f the gate was open, i t entered the kicksorter. Pulses from the side counters were driven by cathode followers into diff e r e n t i a l discriminators whose output was fed into a tr i p l e coincidence mixer c i r c u i t . The third pulse into the triple coincidence mixer was the dynode pulse from the centre channel photomultiplier which had f i r s t been amplified and clipped to resemble the pulses from the side channel discriminators. If a tr i p l e coincidence INPUT PHASE SPLlTTiR V I 6 J 6 100 -o2.0 A M P L I F I E R V 2 6 A H 6 ^ V 3 6 A H 6 V 4 6 A N 5 5oo a w =T=6>F. 0 0 1 C.'.ZZ .001 . 0 t ; 4.7K ZW. .01 H I -CATHODE FOLLOWER. V 5 6 A H 6 IOK WW IOOX 47* 2W • 1 . 0 1 r ' ^ H l - " < r~x GAIN +75V-+7»V. •ww-I O 0 K l > l i < 5 0 O K i i S o * F l 1 >33o 4r »500K .01 VNA/— IOOK tee .001 IOO > 5 0 K X A M PL. V 6 6 A H 6 F L O P 6 A K 5 + 2 2 5 2 . 1 K < 2.1K L N A R R O W C H A N N E L B I A S >20K. I00K 20K IOK t IN5BA - N — T A A A " '4.7K -TV. 2 A K 1 0 0 k 'CARBON LINCAR 22.* CATHODE FOLLOWER B A L A N C E 5 0 K '1R 2 0 0 K ^ »f J. O-IO© I * SOK A A A r l 20O £ + 75 v. — o StT LOWER OISCRIM. :R2 I N 5 8 A -of— <I0K I N 34-^ '4flK 7.5*1-2W. 30 IN58A -i»—H<Hv-I N 5 8 A »5K # I00K : 2 2 K •330K 3 - 2 0 * UPPER •500 DISC. our --2K T O ANTI-COINC. B «108 v . i. .01 - t - 7 5 v . 2 o o ^ V I O 6 A H 6 V I I 6 A H 6 F Y t a P U L S E A M P L I T U D E A N A L Y S E R - D I S C R I M I N A T O R S •Z&OK. f /*N \ | r * < WW. SET UPPER 3>ISC. " V I 2 - 6 A K 5 ."=" V I 3 " 6 A K 5 * N O B L C L O Y Re&ATOH t T R I M ft, R x TO B A L A N C E O U T J>IODE C A P A C I T Y P U C S C . ( B E T W E E N .S" 1 I M ) . _ • . _ . a o u A KORM1L f T R I M TO H A K E 2 0 v = I " " r A N * 2 v r | 0 O r A ON N A R R O W C H A N N E L . PULSE FROM LOWER Disc. (A) 6 8 O o-^ nrvmr\—/vw-l , B 680 .25 j*sec. D E L A Y 2«l2SO\*- 50 K I M •2.2K 10 K WW I N 3 + S • o l C H A N N E L O U T P U T 75 -IM-P U L S E -o FROM UPPER J )/SC. ( B ) T R I P L E COINC. SENS. + 2 2 5 V . /t>—*\ 0 1 ^ C O I N C . H | — * ) 0 U T P V T IOO 1 0 0 K 7,' O N C O I N C . 7 1 3/ I N P U T 1 0 0 x - v C O I N C . -N—z) i N P u r > I O O K +75 v. Fig. 11. P U L S E A M P L I T U D E A N A L Y S E R A N T I C O I N C I D E N C E A N D T R I P L E C O I N C I D E N C E C I R C U I T S f a c i n g F i g . 19 - 33 -occurred, the gate pulse generator was triggered, driving a pulse into the gated biased amplifier to open the gate. The centre counter head amplifier and power supply-circuits were the same design as described i n the last chapter, except that the collector load of the phtotmultiplier was 0.1 Megohms in order to provide a pulse of about one microsecond duration which would be passed undistorted by a gate circuit with only a two microsecond "open" time. The pulse then rose to 90% of i t s value i n 0.1 psec., f a l l i n g to half i t s maximum after 1 jisec. The side counters were similar to the centre counter except that EMI 6262 photomultipliers were used and a separate high voltage supply was used to provide the dynode voltages. These last circuits have been described by Griffiths (1953). The di f f e r e n t i a l discriminator c i r c u i t i s shown i n Fig. 10. V2, V3, and lk amplify the input pulse5 then inject i t at the gfids of the cathode followers V5 and V10. The D. C. voltage levels of the cathode followers are set by the Helipot discriminator potentiometers so that the amplifiers V6 and V l l do not respond to the input pulse u n t i l the pulse voltage reaches the discrimination levels. If the pulse height i s greater than the level of the lower discriminator, the amplifier causes the " f l i p - f l o p " c i r c u i t , V7 and V8, to trigger. If the pulse is also greater than the upper discrimination level, the "f l i p - f l o p " tubes..¥12 and W13 i n i t i a t e a pulse. Fig. 12. GATE PULSE GENERATOR - 3^ -* The discriminator output pulses are fed into the anticoincidence circuit of Fig. 11. The lower discriminator pulse is delayed i n order to compensate for the extra time that the pulse into the discriminator takes to reach the upper discrimination level. If only the lower discriminator f i r e s , i t s negative pulse switches the cathode current of V6,-previously carried entirely by the l e f t side of the tube, onto the right side, resulting i n the pulse being reproduced at the channel output. If the upper discriminator also f i r e s , the negative pulse i t applies to the grid of the right side of $6 keeps this side from conducting, and no output pulse appears. In the triple coincidence c i r c u i t , F i g . 11, the l e f t hand grid of Ylh is biased at a lower voltage than that of the right hand grid and both the grids of Vl5. Only i f negative pulses appear at a l l three of the latter mentioned grids w i l l the common cathode current be switched to the right hand side of Ylh to produce a "coincidence output" pulse. The gate pulse generator c i r c u i t , triggered by the tri p l e coincidence pulse, is shown i n F i g . 12. Here VI i s an', ordinary univibrator i n which the trigger point i s set by adjusting the D. C. voltage level on the grid of the right hand half of the tube. Pulse discrimination i s necessary because the triple coincidence cir c u i t of Fig. 11 produces small pulses i f one or two pulses appear at i t s inputs. The two microsecond pulse from VI, Fig. 12, is amplified to 30 volts, positive polarity, and driven into the biased amplifier gate cir c u i t by the.cathode follower V3. GATE PULSe AMP. sic. IN <->(7n»-CoiNcoeKeeL CIRCUIT HISCRIM-C A T H O f t t P O U U • 2 9 5 v. - v / W 3 0 K I O O K * N O B L E L O Y Fig. 13. G A T E D B I A S E D A M P L I F I E R - 35 -The gated biased amplifier cir c u i t is shown i n F i g . 13. If the switch SI i s on "coincidence", the gate pulse, amplified by VI and V3, is applied i n negative polarity to the grid of Vh, The centre channel pulse from the Northern E l e c t r i c amplifier is applied to the grid of V5. The D. C. voltage level of the grids of V*f and V5 i s set by the "Helipot" d i s c r i -minator potentiometer. If no gate pulse appears at the grid of V*f, the centre channel pulse merely shuts off the current i n V5 and transfers i t to V*+, but i f the gate pulse should appear on the grid of V*+ just before this, i t w i l l switch the current from Vh to V5, and V5 w i l l act as an ordinary cathode follower to the centre channel pulse. The biasing action of the amplifier depends on the discriminating action of V6 and V8. The grid of V6 i s normally at earth potential, so that the tube current i s shut off. It w i l l not conduct current u n t i l the cathode potential of V5 reaches earth potential. As soon as the centre channel pulse has driven the cathode voltage of V5 to this level, V5 w i l l begin to conduct, and the current i t carries w i l l cause the potential of the grid of V8 to f a l l since a. condenser connects this grid to the plate of V5. The cathode potential of V8 w i l l then follow that of i t s grid, and since this voltage is fed to the grid of V6 through an attenuator network, V6 w i l l tend to be shut off. Since V8 has a very high transconductance, i t w i l l make the grid potential of V6 follow that of the cathode of V5 exactly, so that the section of the pulse on the cathode - 36 -of V5 which is less than earth potential w i l l he reproduced and amplified at the cathode of V8. The last tube, V12, is a cathode follower which drives the pulse on the cathode of V8 into the kicksorter. On i t s grid i s a pulse stretcher, which extends the pulse to approxi-mately 10 microseconds duration. Because the energy selective circuits i n the side channels could not discriminate u n t i l the input pulse reached i t s maximum height, the "channel output" pulses were produced. 0.5 yuisec. after the input pulse began. Thus an equal delay was. required for the third pulse into the triple coincidence mixer. A one foot length of 1000 ohm inductive delay cable was placed between the centre counter cathode follower output and the ring of the three amplifier to create such a delay. In order to place the centre channel pulse i n the centre of the gated interval, a 1.5 J*sec. delay was needed to compensate for the delays i n the side channels and'the time required for the gate pulse to rise to i t s maximum. Two feet of the delay cable just mentioned, placed between the head amplifier and the Northern Electric amplifier provide the required delay. Since i t should be possible to obtain an energy resolu-tion equivalent to that of a single crystal f u l l energy peak, such resolution was made the goal of the research. Other arrangements of the apparatus of Fig. 9 were tried, as well as FLATE I I . D i s p o s i t i o n of the counters, s h i e l d i n g removed. - 37 -the use of fast coincidence mixing of the side channel pulses. The added complexity of the latter added l i t t l e to the back-ground rejection i f the side counters were well shielded, and was dropped i n favour of four inches of lead shielding. It was thus impossible to put the centre crystal face closer than 13 cm. from the target, and this distance was used i n the experiments. The limitation on the solid angle of the spectrometer v was not serious because the photomultiplier gain drifted i f the counting rate was made too great. The photomultiplier used in the centre counter was much more stable i n gain than any of the others tested, and could tolerate counting rates for pulses representing over 1 Mev. energy release i n the phosphor as high as 5»000 per second. Plate I is a photograph of the apparatus. The counters are shielded with lead. Plate II shows the disposition of the counters without the lead shielding. In the actual experi-mental work, the centre counter photomultiplier was horizontal; the vertical position, however, helps to preserve the optical seal of the crystal to the photomultiplier. 3. Operation and Results The centre counter photomultiplier was operated at 1500 volts, the side counter photomultipliers at about 1600 volts, adjusted to make the side counters have equal gain. The power to a l l equipment was l e f t on continuously to obtain maximum day to day st a b i l i t y . - 38 -The energy discrimination levels of the side channels were set by viewing a spectrum of the annihilation radiation from a Na 2 2 source with a "Tektronix" oscilloscope receiving pulses from the collector' of the side channel photomultiplier. With the oscilloscope trace triggered by the output from the side channel receiving the corresponding dynode pulse, the discrimination levels were adjusted u n t i l only the bright f u l l energy group of pulses was visible on the screen. The same person could reproduce the voltage settings every time to at least 5$j and since the settings were not c r i t i c a l , being at "valleys" i n the spectrum, a reproducible efficiency was ensured. The procedure also lent i t s e l f to quick checks on sta b i l i t y . Calibration of the gamma ray energy-pulse height scale for the spectrometer was done i n the usual fashion. A standard pulse generator, described by Griffiths (1953)> injected pulses at the grid of the centre counter head amplifier cathode follower, the gate was disabled, and the standard pulse height varied u n t i l the slight fluctuations i n pulse height made the kick-sorter count equally in two neighbouring channels. Single and three crystal spectra of the 1.28 Mev. gamma ray from Na 2 2, the 2.62 Mev. gamma ray from ThC", and the 6.1^, 6.91, and 7.12 Mev. gamma rays from the reaction Fl9(p,oiy)ol6, then provided the calibration. The 17.6 Mev. gamma ray from the reaction Li'7(p,a')Be^ was used to establish the linearity of the energy scale, but was not used regularly because i t s spectrum does not C H A N N E L N U M B E R -,39 -provide a well defined energy measurement. The standard pulses were injected at the grid of the cathode follower i n order to include a l l the amplifiers and other electronics i n the calibration. It was very d i f f i c u l t to avoid gain drif t s from excessive counting rates i n the centre crystal. To keep this rate at a low enough level, the total counts i n one side channel counter were monitored and the beam current altered to keep this rate constant. After long neutron bombardment, there was a considerable activity created i n the crystals from neutron capture, with a period of.roughly two hours• Aside from the count rate effect, some correlation between gain drif t s and ambient temperature of the photomulti-plier was noted. Care was taken to keep the temperature from varying. The f i n a l results can be judged from Figs. l*f and 15. Fig. Ih shows a spectrum of the radiations from the reaction F 1 9(p Joty)o l D at Ep = 873 kev. The large peak has % energy resolution, occasionally h% was achieved. The spectrum shows the expected " t a i l " from the escape of Bremsstrahlung from the fast pair electrons and from the escape of the electrons them-selves. The " t a i l " contains about half as many pulses as the main peak. The dip at the base of the peak is reproducible and can be explained by the differ e n t i a l absorption of the softer Bremsstrahlung, as mentioned i n the last chapter. (See Fig. 3.) 1.60 M E V . ** O o H J \C l«f lis l» ZU "dA Zt C H A N N E L N U M B E R Fig. 15 Three crystal (Curve A) and j i n g l e crystal (Curve B) pulse height distributions from ThC" radiation showing resolution. Curve A has been shifted along the abscissa to. i l l u s t r a t e background rejection. - 1+0 -Fig. 15 shows single and three crystal spectra of the 2.62 Mev. gamma ray from ThC". Here no " t a i l " i s present and the effect of side channel discrimination best estimated i n i t s rejection of background. The single crystal spectrum illustrates the resolution of the counter that was achieved through the use of careful ©ajyataUt:;mounting technique. The efficiency of the spectrometer has not been measured, but is estimated to be about 10-3 for 6 Mev. radiation. An increase i n efficiency of almost a factbor 5 could be obtained i f the side channel crystals were twice as long so that the f u l l gamma ray energy group of pulses i n the pulse height d i s t r i b -utions of the side counters contained half the total number of pulses. Apparently the photomultiplier used i n the centre channel must be carefully selected for sufficient gain s t a b i l i t y to withstand high counting rates. 3-09 MEV. CHANNEL NUMBER 3 0 0 F i g . 17. Three crystal spectrum of the radiations from B 1 0 +• d, E d » l.h Mev. 2 5 0 1 ^ 2 0 0 J Z < CJ UJ (J) o o 1501 loop 6 - 4 9 MEV. 5 0 1000 _L PULSE HEIGHT - VOLTS 1 2 - 2 8 1 4 - 4 8 -4 I 1 6 4 3 _L 18-76 _1 T 1 1 ' i 1 1 1 1 1 *~r 1 1 r— 4 6 8 1 0 12 14 16 18 2 0 2 2 2 4 2 6 2 8 CHANNEL NUMBER T 2 2801 240 ID Z200I < X o Q . CO § 1201 O O 8 0 h 4 0 k 6-52 MEV. Fig. 18. Three crystal i spectrum of the radiations f from B 1 0 d, Efl » 1.^ Mev. 7-29 MEV T 2 14-00 _L T 4 PULSE HEIGHT-VOLTS 16-27 18-47 - J L I T 1 1 1 1 r 8 10 12 14 16 18 20 CHANNEL NUMBER 20-61 —'T™ 22 22-75 "I L T " 26 28 24 2 4 6 8 IO 12 14 16 18 20 22 24 26 28 CHANNEL NUMBER CHAPTER IV GAMMA RADIATION FROM THE DEUTERON BOMBARDMENT OF BORON TEN The bombardment of Bl° with deuterons initiates the reactions B^CdjpoOB 1 1 and B ^ C d j n O C 1 1 , and thus provides an opportunity to study the radiative transitions between the levels of the minor nuclei B 1 1 and C-1-1 simultaneously. Accordingly, the radiations from BlO -f- d were observed with the three crystal pair spectrometer just described. Five bombarding energies were used: 0.8, 1.0, 1.M-, 1.7 and 2.2 Mev. A set of representative spectra, obtained with the three crystal spectrometer for Ed = 1.*+ Mev. are shown i n Figs. 16, 17, 18, and 19. Here the peaks are labeled with the f u l l gamma ray energy. They show the great resolving power of the spectro-meter. A description of the experimental procedure used to obtain these and the other spectra, and their interpretation, has been given in Mr. Sample's thesis (Sample 1955) and w i l l not be entered into here. - I f 2 -APPENDIX The mounting of sodium iodide crystals as done i n this laboratory has been described by Azuma (1953)» but certain details of the process are worth underlining. It has been a major problem to keep the deliquescent crystals unclouded by hydrolized material on the surface over long periods of time. It was found that lucite windows seem to contain enough water to cloud the crystal, and rubber gaskets are inclined to leak. Therefore, thin glass windows were used, sealed permanently to the aluminum case with "Araldite" hot setting cement. Glass absorbs more ultraviolet light than lucite, but good photo-graphic plate or Corning 77*+0 pyrex were tested with a comparator spectrograph and found sufficiently good. Ultimately, quartz may be used. The l i d to the case was threaded; silicone sealing compound on the threads and Duco enamel over the outside made a permanent seal. The threaded l i d also held the crystal firmly against the window and spread the million centistoke silicone o i l optical seal thinly and evenly. The crystal was polished with a mixture of equal parts of n-butyl alcohol and xylene absorbed in a blotter. The crystal face to be polished was wiped free of the petrolatum protecting i t from moisture, rubbed i n the blotter, quickly dipped i n pure xylene to neutralize the alcohol, then returned to the petrolatum bath. In the dry box, reflecting surfaces were roughened with sandpaper, and very dry magnesium oxide powder was rubbed into - ^3 -them u n t i l i t clung of i t s e l f . The clear surface was sealed to the window with the heavy o i l which had previously been placed i n situ under vacuum to remove a l l small air bubbles. - kh -REFERENCES Alexander, T. K.,'1955, M. A. Thesis, University of B. C. Ajzenberg, F., and Lauritsen, T., 1955, Rev. Mod. Phys., 2Z, 77. Aron, W. A., Hoffman, M. M.. and Williams, C , 19^ 9, At. En. Com. publication #663. Azuma, R. E., 1953, M. A. Thesis, University of B. C. Bair, J. K., and Maienschein, F. C , 1951, Rev. Sc. Inst., 22, 3^3. Brown, A. B., Snyder, C. W., Fowler, W. A., and Lauritsen, C. C , 1951, Phys. Rev., 82, 159 Caldwell, R. L., and Turner, S. E., 195L<-> Nucleonics, 12, #12, Day, R. B., and Huus, T., 195*+} Phys. Rev., 2l9 1003. Davisson, C. M., and Evans, R. D., 1952, Rev. Mod. Phys., i k , 79. Edwards, M. H., 1951, M. A. Thesis, University of B. C. Grif f i t h s , G. M., and Warren, J. B., 1950, Proc. Phys. S o c , i + i , 1050. G r i f f i t h s , G. M., 1953, Ph. D. Thesis, University of B. C. Heitler, W., 195^, Quantum Theory of Radiation, Third Edition, Clarendon Press, Oxford. Johansson, S. A. E., 1952, P h i l . Mag., }£, 2^9. Krone, R. W., and Seagondollar, L. W., 1953, Phys. Rev., ,22, 935. L i , C. W., Whaling, W., and Fowler, W. A., 1951, Phys. Rev., M> 512. Mandl, F., and Skyrme, T. H. R., 1952, Proc. Roy. Soc. (London), 212, *+97. Masden, C. D.,.1953, Dan. Vid. Sels. Medd., 2Z, #13. Mclntyre, J. A., and Hofstadter, R., 1950, Phys. Rev., 28, 619. McClelland, C , Mark, H., and Goodman, C , 1955, Phys. Rev., 21, 1191. - U£ -Paris, C. H . , and Endt, p. M . , 195*+, Physica, 20, 585. Richards, H. T., Smith, R. V., and Browne, C. P., 1950, Phys. Rev., 80, 52^. Roberts, P. W., 1953» Proc. Phys. S o c , A66, 192. Sample, J. T., 1955* Ph. D. Thesis, University of B. C. Sharp, ¥. T., Kennedy, J. M., Sears, B. J., and Hoyle, M. G., At. En. of Can. Ltd., Publication #97. Walker, R. L,, 1950, Phys. Rev., 22j !72. Weisskopf, V. F., 1951, Phys. Rev., 82, 1073. West, H. I., and Mann, L. G., 195^ > Rev. Sc. Inst., 2£, 129. Wilkinson, D. H., 1953, P h i l . Mag., }±k, h$0. Woodbury, H. H., Day. R. B., and Tollestrup, A. "V., 1952, Phys. Rev., 8£, 760. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items