30Jun'48 UNITED STATES ATOMIC ENERGY COMMISSION N CLOUD-CHAMBER ENERGY MEASUREMENT OF PHOTONEUTRON SOURCES by D, J. Hughes C, Eggler Argonne National Laboratory This document consists of 8 pages. Date of Manuscript: June 6, 1947 Date Declassified: July 9, 1947 This document is issued for official use. Its issuance does not constitute authority to declassify copies or versions of the same or similar content and title and by the same author(s). Technical Information Division, Oak Ridge Directed Operations Oak Ridge, Tennessee MDDC - 1082 12_a2S-Co»er CLOUD-CHAMBER ENERGY MEASUREMENT OF PHOTONEUTRON SOURCES By D. J. Hughes and C. Eggler ABSTRACT The energy distributions of the neutrons from several photoneutroi-' sources have been measured by means of the range distribution of recoil protons produced in a hydrogen-filled cloud chamber. It is shown that slowing down in the source itself produces a broad spread in energy instead of the al- most monoenergetic spectrum to be expected from a single y energy. The values of the mean neutron energy, E, and the maximum energy, Ej^-, for the sources measured are as ff'Iows: Source E (Kev) E^ (Kev) Na-Be 800 1020 Na-DjO 220 320 Mn-Be 300 375 <150 <150 In-Be 300 375 <150 <150 Sb-Be 35 68 As the Ej. values should be directly related to the energies of each of the >• sources, they are com- pared to the y energies as reported by different observers for each source. INTRODUCTION The advent of chain-reacting piles has made available intense gamma sources which can be used to produce photoneutrons by the photodisintegration of the deuteron or of beryllium. Such photo- neutron sources have proved to be extremely useful and convenient neutron sources in the region 30 Kev to 1 Mev. If only a single gamma ray above the photodisintegration threshold is present and the amount of deuterium or beryllium is very small, then the emitted neutrons show pracf ically no spread in energy. Such a monoenergetic neutron source is, of course, much superior to the usual Ra (a) Be source with its great spread in neutron energy. Actually, in order to obtain sulficient in- tensity, the amount of deuterium or beryllium which must be used is large enough so that significant moderation of the neutrons occurs by elastic scattering and a broad band of energies results. During the year 1944, Wattenberg' investigated the production and calibration of photoneution sources, using various gamma emitters produced in the Argonne pile. The gamma sources were sur- rounded by amounts of deuterium or beryllium about one centimeter in thickness in order to obtain sufficient intensity, and as a result the nearly monoenergetic neutrons emitted in the photodisinte- gration were slowed down appreciably in the surrounding material. Wattenberg estimated the averasio MDDC - 1082 ( 1 2 ] MDDC - 1082 energy of his sources by measuring the average scattering cross section of hydrogen for the neutrons of each source. The energy corresponding to the observed hydrogen cross section was then obtained from the curve of Bohm and Richman.^ The energies as determined from the hydrogen scattering were always less than that expected from the gamma energy, thus showing that the moderator did have a definite effect in lowering the average neutron energy. In order to aid in the understanding of the actual energy distribution of the neutrons emitted by the sources, it was decided to measure the neutron energies by means of recoil protons in a hydrogen- filled cloud chamber. In addition, it was hoped that investigation of the actual energy distribution would show whether some of the sources consisted of several neutron groups instead of a single one. Some of the gamma sources had been reported as having more than one gamma energy above the photodisintegration threshold, but it was impossible to ascertain from the average energy, as meas- ured by the hydrogen scattering, whether miiltiple groups were present or not. METHOD The main difficulty in the measurement of the neutron energies with the cloud chamber is due to the fact that, while the sources are only moderate in neutron strength, they are intense in gamma activity. Thus a source strong enough to give one recoil proton per expansion when placed about two feet from the chamber must be several curies in gamma strength. Such an intense gamma activity near the chamber makes it rather difficult to obtain clear proton tracks. It was found, however, that if the photoneutron source was surrounded by several inches of lead to reduce the gamma activity to some extent, and if the expansion ratio of the chamber was kept quite low (which emphasizes proton over beta tracks), it was possible to obtain quite sharp contrast between the protons and the intense background of beta tracks. The presence of the lead does not change the neutron energy spectrum appreciably, because energy losses for elastic scattering with lead are small and inelastic scattering does not take place at photoneutron energies. The apparatus is shown diagrammatically in Figure 1. It was placed on supports about 6 feet above the floor, in a large room, to reduce the scattering of neutrons from walls and floor. The gam- ma source could be lowered by remote control into a large lead pot on the floor while the chamber was being adjusted, film changed, etc; then it could be replaced in operating position and several hun- dred pictures taken without the necessity of approaching the apparatus. The photoneutron source it- self is of the same construction as those described by Wattenberg, that is, a tube 2 cm in diameter and 5 cm long containing the gamma source, located in a cylinder of beryllium or a deuterium -filled cylindrical can 3.8 cm in diameter and 5.1 cm long with a 2.2-cm diameter axial hole. The cloud chamber, 30 cm in diameter, is very similar to that described by Jones and Hughes.' For most of the sources measured it was filled with hydrogen gas and water vapor to a pressure of 82 cm of mercury. The stereoscopic pictures were taken with a standard mirror arrangement using 35-mm Eastman XX film and an f 3.5 lens. The light source was a xenon-filled capillary, flashed by discharging a 50-/if condenser bank at 2000 volts. The xenon lamp, a General Electric "Flashtube FT 26," has proved to be an extremely convenient cloud-chamber light source. The lengths of the recoil protons and their direction of motion relative to the incident neutrons were determined by a stereoscopic reprojection of the negatives. The proton ranges in standard air were obtained by comparison with the observed range in the chamber of the alphas from a plutonium source mounted in the chamber. In the comparison, a correction was made for the fact that the stop- ping power of the chamber for alphas relative to protons is a function of the proton range. The proton range in air was converted to energy, using the range energy curves given by Livingston and Bethe* and the neutron energy calculated from the proton energy and the angle, 6, between the neutron and proton direction. The energy spectrum of the neutrons from the source could then be plotted from the observed numbers of recoil protons as a function of energy. A correction must be made, of course, for the effect of the change in the scattering cross section of hydrogen with neutron energy, but be- cause of the nearlv monochromatic nature of the sources such a correction is small. l2-229-p2-bu MDDC - 1082 [S Cloud Chamber Pu Alpha Source -30 cm Photo- Neutrorv Source 65 cm cm Figure 1 "Q Lead Shield Be or DaO Cylinder y Ray Source Brass Tube Lead Coffin for Source Storage RESULTS If no slowing down takes place in the deuterium or beryllium of the source itself, then the energy spread in the emitted neutrons should be very small. The energy spread in this ideal case is caused by the difference in direction of the neutron and the gamma ray and is given by < = E y cos / 2(A-l)(Ey-Q y ^l 931 A' / where 6 is the energy spread, Ey the gamma energy, A the mass of target nucleus, and Q the thresh- old energy. The value of this energy spread is usually of the order of one per cent, which is much less than the energy spread caused by slowing down in an actual source. Because of the low atomic weight of deuterium and beryllium, they are excellent moderators and distort the neutron energy spec- trxun even when present in small amounts. The resulting energy spectrum is e3q)ected to have a max- imum neutron energy corresponding to that given by the gamma energy but an average energy some- what lower. The actual energy spectra were measured for several photoneutron sources of interest, and the findings will be discussed for each source. Na-Be The spectrum obtained from 106 recoil protons which were within 30° of head-on collision is shown by the diagram ofTigure 2. It is seen that the distribution, while showing no great spread in energy similar to a Ra Be source, is by no means monoenergetic. The spectrum is, of coxirse, dis- torted to some extent by errors of measurement. However, if only those recoils are chosen which are within 20° of head-on, the spectrum does not narrow appreciably, the width at half maximum, 4] MDDC - 1082 which is 33% of the maximum energy in Figure 2, changing only to 31% in the 20° case. The true width is probably slightly less than the latter value, say about 25%. 20 N 15 10 5- J] Na-DzO J\ Na-Be \I Ln n Em I 0.2 0.4 0.6 0.8 1.0 1.2 E, Mev Figure 2 The most probable neutron energy from Figure 2 is about 825 Kev, while the arithmetic mean energy is 800 Kev, Wattenberg finds a mean energy for this source from the mean hydrogen scatter- ing cross section of 830 Kev. It seems then that the spectrum of Figure 2 is a good representation of the actual neutron distribution of a source, made according to Wattenberg's method. The maximum energy in the spectrum can of course be identified with those neutrons which are unmoderated and which should therefore correspond to the energy calculated from the Na gamma en- ergy. The Na gamma energy has recently been measured as 2.76 Mev by Siegbahn,^ in agreement with an earlier measurement of Elliot, Deutsch, and Roberts.^ Other measurements have given values for the energy ranging as high as 2.94 Mev. The maximum neutron energy to be expected, assuming Ey to be 2.76 Mev, is marked as En^ in Figure 2, Considering the inevitable straggling of the experi- mental points, the agreement between Ej^ and the observed upper limit is quite satisfactory. A gam- ma ray has been reported of energy higher than the 2.76-Mev gamma, but no tracks were found in the present spectrun^ (or in the Na-D20 spectrum) to indicate the presence of such a gamma ray. If it exists, it does not contribute an appreciable number of neutrons in the photoneutron source. Na-Q,0 The spectrum based on 75 recoils is shown in Figure 2. The theoretical maximum, based on a 12.233.p4.bu MDDC - 1082 [ 5 gamma energy of 2.76 Mev and a D^O threshold of 2.18 Mev, is indicated, and it is seen that the max- imum neutron energy corresponds quite well with the theoretical value. The most probable value of the distribution and the average are at about 220 Kev (70% of Ej^), which is the same value as the mean energy which Wattenberg finds for this source by the hydrogen cross section method. The width of the energy distribution at half maximum for the Na-D^O source is 20% of Ej^^. Mn-Be Gamma-ray measurements'''^ for Mn'* had shown two gamma rays of energy 1.81 and 2.13 Mev. , As both these energies are higher than the threshold in Be, one would expect two groups of photo- neutrons. Wattenberg actually found from the scattering cross section of hydrogen that the photo- neutrons he observed were due to a gamma ray of energy 1.83 Mev. (He also found an extremely weak group of photoneutrons in deuterium which would indicate a 2.7-Mev gamma). As his method of estimating energies from the hydrogen scattering cross section gives only the mean energy of the neutron group, it was impossible for him to say if there were any neutrons from the 2.1-Mev gamma. The energy spectrum of the Mn-Be neutrons was measured in the cloud chamber to determine if only one group of the neutrons was present, as seemed likely from Wattenberg's result, or if two groups were present, as would be indicated by the gamma energies. The results are shown in the lower curve of Figure 3. It was found that two groups of neutrons were definitely present (no effort was made to study the small number of neutrons which would be caused by the 2.7-Mev gamma). The energy of the most abundant group was too low to be measured accurately, as the recoil protons were of such short range. The higher energy group has an average energy of about 300 Kev, and it is pos- sible to estimate the energy of the gamma ray causing the group. The maximum energy Ej^ is chosen as 375 Kev, by taking E-^ slightly less than the apparent maximum energy in analogy with the spectra of Figure 2. A value of 375 Kev for Ej^ then gives 2.05 Mev for the gamma-ray energy. This energy is somewhat less than the earlier value 2.13 Mev but agrees extremely well with a recent determina- tion of 2.06 Mev for this gamma ray by Siegbahn.^ It is definite then that the 2.06-Mev gamma pro- duces photoneutrons in addition to the 1.81 -Mev gamma. The relative numbers of photoneutrons in the low and high energy groups are about 90% and 10%, respectively, as determined by counting recoil protons and correcting for the change of hydrogen cross section with energy. Because of the low intensity of the high energy group, it was not indicated as a discrete group by Wattenberg but it probably increased his average energy slightly. Thus the 1.83 -Mev gamma-ray energy that he inferred is probably high for this reason. Siegbahn's^ recent determination of the low energy gamma is 1.77 Mev. In-Be Indium is of doubtful value as a gamma emitter for photoneutron sources because of its short half-life (54 minutes). However, if only one group of neutrons were present, then the difficulty caused by the short half-life would not be insurmountable. Gamma-ray measurements had indicated energies of 1.8 (spectrometer) and 2.3 Mev (cloud chamtar) for those higher than tne Be threshold, so it was decided to measure the photoneutron spectrum to see if several groups were actually present. The spectrum obtained, shown in the upper part of Figure 3, contains two groups of neutrons, one of mean energy about 300 Kev and a group of energy too low to be measured accurately (about 100 Kev). The higher energy neutron group comprises 59% of the total and its E^ for Na indicates a 2.1-Mev gamma from indium. The low energy group cannot be measured accurately, but it indicates a gamma of rough- ly 1.8 Mev. It seems, therefore, that both energies are present in indium and in such intensities that they give roughly comparable groups of photoneutrons. Because of the presence of the two neutron groups, the value of In Be as a photoneutron source is much reduced. Sb-Be The highest energy gamma ray from Sb has been reported'°>'^ as having energies ranging from l2-223-pS-bu 6] MDDC - 1082 N 15 10 41% In-Be 59% .1 .2 .3 .4 .5 Mev J Mn-Be 15 10 "L 5 90% 10*/ - h 1 .3 A Figure 3 .5 Mev 1.70 to 1.82 Mev. This discrepancy in the gamma ray is large enough so that the resulting discrepancy in the photoneutron energy is quite serious. Shariff-Goldhaber and Kiaiber^^ measured the energy of the photoneutron from an Sb-Be source and found Ej^^ to be 115 Kev, which would indicate a gamma energy of about 1.75 Mev. Wattenberg's value for the average energy of the photoneutrons is 24 Kev, from which he obtains a gamma energy of 1.67 Mev. The spectrum of photoneutrons was studied with the cloud chamber mainly to investigate this rather large discrepancy. Because the neutron energy is so low, the cloud chamber was operated at the lowest possible pressure, to increase the range of the protons to a measurable value. The chamber could be operated with hydrogen at a pressure of about 8 cm, using water as the vapor. Under such conditions the range is about 25 times the range in air, and a 20-Kev proton, which would have a range of only 0.3 mm in air, will have a range of almost a centimeter in the chamber. The source strength of Sb-Be is very low, so it was possible to obtain oiily 20 recoil protons, caused by nearly head-on collisions. The distribution obtained is plotted in Figure 4. In spite of the la-aas-p*-*" MDDC - 1082 [7 extremely small number of tracks, it appears that a single group of neutrons of average energy of about 35 Kev is present. This value is in rather good agreement with the mean energy of the photo- neutrons of 25 + 15 Kev measured by Wattenberg. Because the neutrons are of such low energy, it should be possible to use the photoneutron energy to determine the gamma energy rather accurately. Unfortunately, the straggling at the upper end of the measured spectrum of Figure 4 is quite large, and Ej^ cannot be determined very accurately. However, E-^ seems to be within the range shown, that is, 68 ± 11 Kev. The value of the gamma energy resulting from this Ej^ is 1.707 1 .012 Mev. The error in this determination of the gamma ray is actually somewhat greater than 12 Kev, because the error in the photoneutron threshold should be included, as well as errors in the determination of the stopping power of the chamber gas. Inclusion of these two errors would probably increase the error in the gamma energy to about 20 Kev. 1 i i Sb -Be C. ^ l_ m L 1 1 J 1 5- 30 40 50 60 70 80 Kev Figure 4 The gamma energy determined from the present experiment agrees very well with the latest direct gamma energy determination of Kruger and Ogle, which gave 1.70 i 0.02 Mev. It is somewhat higher than the value 1.67 Mev which Wattenberg obtains from the mean energy of the photoneutron source. i2.aas-p7'bu 8 ] MDDC - 1082 ACKNOWLEDGMENTS We wish to extend thanks to Dr. A. Wattenberg for his help in preparation of the sources used and for valuable discussion, to T. Brill for the design of cloud-chamber control circuits, and to N. Goldstein and S. Joshemski for aid in taking data. REFERENCES 1. Wattenberg, A., Phys. Rev. 71, 497 (1947). 2. Bohm, D., and C. Richman, Phys. Rev. 71, 567 (1947). 3. Jones, H., and D. Hughes, Rev. Sci. Inst. 11, 79 (1940). 4. Livingston, M. S., and H. A. Bethe, Rev. Modern Phys. 9, 245 (1937). 5. Siegbahn, K., Phys. Rev. 6. Elliot, Deutsch, and Roberts, Phys. Rev. 67, 273 (1945). 7. Deutsch, M., and A. Roberts, Phys. Rev. 60, 362 (1941). 8. Elliot, L. G., and M. Deutsch, Phys. Rev. 63, 321 (1943). 9. Siegbahn, K., Ark. Mat. Astr. Fys. 33A (2), Paper 10 (1946). 10. Kruger, P. G., and W. E. Ogle, Phys. Rev. 67, 273 (1945). 11. Mitchell, Langer, and McDaniel, Phys. Rev. 57, 1107 (1940). 12. Schariff-Goldhaber and Klaiber. Phvs. Rev. 61, 733A (1942). ^ERSITY OF PLOR'DJ,,, .^ iiiiliill"*