. • . 14 | OF ORNL . i : of . < is ; . . ľ : fo:.. de . 1 G . II16 . li . i 1 - - SO HWIN II 3.6 I IN 4 . PRI" 2. ||125 | 14 LIS : 11 . 1 į A MICROCOPY RESOLUTION TEST CHART NATIONAL BUREAU OF STANDARDS - 1963 ,, res WY Y . LEGAL NOTICE This report was prepared as an account of Government sponsored work. Neither the United States, nor the Commission, nor any person acting on behalf of the Commission: A. Makes any warranty or representa- tion, expressed or implied, with respect to the accuracy, completeness, or usefulness of the information contained in this report, or that the use of any information, appa- ratus, method, or process disclosed in this report may not infringe privately owned rights; or . B. Assumes any liabilities with respect to the use of, or for damages resulting from the use of any information, apparatus, method, or process disclosed in this report. As used in the above, “person acting on behalf of the Commission”includes any em- ployee or contractor of the Commission, or A GV Et EL . . A . that such employee or contractor of the Commission, or employee of such contractor prepares, disseminates, or provides access to, any information pursuant to his employ- ment or contract with the Commission, or his employment with such contractor. U LLIU... W .... .*77- VW .... ..- 4 .-..- .-. *T* VETT. TT.", PEITT * (Presented - Symposium on Pulsed High Intensity Fission Neutron Sources, Washington, D. C., February 18, 1965) ORN P-1116 CONF -(050217-). NEUTRON CROSS SECTION MEASUREMENTS USING PULSED SOURCES * J. A. Harvey Oak Ridge National Laboratory - . . . . 1. Introduction When Lowell Bollinger called me two weeks ago and asked if I would substitute for him for this talk today, I agreed to do it if he would give mu his opinions on the subject. He had looked into some of the published work from the IBR, the USSR pulsed reactor at Dubna, and had come to the following conclusions: . (1) That the USSR pulsed reactor was not a particularly suitable source for neutron cross section measurements in the resonance energy region because the pulse width of ~ 40 uses was too large and this wide pulse width demanded long flight paths and low repetition rates to obtain reasonable resolutions. That a mechanical chopper situated at & pulsed reactor with a high peak power" (much greater than the average powers now available from research and a high pulse repetition rate present steady-state reactors), would obviously be superior to existing fast-chopper spectrometers and would have lower backgrounds which would be an important advantage in some experiments. - LEGAL NOTICE - et at the or costruct sporord work, Wattbar do Ouled ottond b duo report or , sort Coquidos, por as porta attoo bobl of the Coontadon: wywraty ortoprenotatine, aprand or lapuede mo rupect to De Rocy- ole Countddon" onclude my *- a Hasution with respect to the wol, or for draugu retty troa the tod, or grocon disclosed la to report The roport wuo popared um Account of Governo of uy tetorantiam, apparatus, method, or propose declared to toda roport may not tatring falaus of the balarendo ployee or contractor of the Commandou, or muxplogue of reach cotructor, to the extent Roch employee or contractor al de Commission, or maplogue at we wodtructor prepare I hit the Dom, pumua icttisau b doncentrate, of prorde accu to, way torcedor permat to do a plot Wed the Cocaminadon, ar bun raplogant with mea contractor. ons, or of my taformation, apparibus, n patrately owood out, or B. A msy aan As A m w way warranty or reported the balorando contains the date report say! I would like to outline some of the parameters which determine the 2016 value of a pulsed neutron source for low energy neutron spectroscopy in the energy range from ~1 to 10° eV in order to compare a spectrometer consisting of a pulsed reactor with a chopper to a spectrometer consisting of a pulsed accelerator with a pulsed multiplier. * Research sponsored by the U. S. Atomic Energy Commission under contract with the Union Carbide Corporation. 2001. PATUT OLEARANCE QSTAINED. REPLASE I THE O IS APPROVED. PROCEDURES ARI NA A AGERWING SECTON. 2. Resolution, Intensity and Background The usefulness of a low energy neutron spectrometer is determined by Its resolution, intensity and background. Resolutions of time-of-flight neutron spectrometers have improved by two orders of magnitude in the past. 10 years and transmission measurements have been made up to 4 keV with an energy resolution of 4 eV which is less than the Doppler width of the resonances. However, at 100 keV the resolution is about 25 times the Doppler width of the resonances and, hence, fine details in this energy region, which is of Interest to the fast reactor breeder program, cannot be obtained with the best present-day spectrometers. For partial cross section measurements at the lower neutron energy, more-intense pulsed sources are needed to make measurements such as, the gamma-ray spectra from individual resonances with a high resolution Ge gamma-ray spectrometer, the angular or mass distributions of f'ission fragments from individual resonances, etc. Almost all experiments are troubled with backgrounds to some degree and more intense, pulsed neutron sources would often improve the signal to background ratio. 3. Neutron Pulsed Source for Time-of-Flight Spectrometer The time-of-flight technique consists of the pulsed neutron source, the flight path, a sample, an appropriate detector and a multi-channel analyzer. All parts must be pushed to their limits, but I will be concerned only with the pulsed source in this talk. If we for the moment assume that only the neutron source determines the width of the neutron pulse, then ****** rk' P' fRk (Pfw'] R (1) nomas -- where N is the number of neutrons of energy E per sq cm per second per resolution width at the end of a flight path_to obtain a resolution R p' is the number of usable . neutrons of energy E leaving the source per second during the pulse f is the pulse repetition rate in pulses per second R is the resolution of the spectrometer, usually expressed in nanoseconds per meter 4 El-Ita ev) = 2.8 x 10-3 R(E(in ey)312 wi is the time width of the neutron pulse of energy E usually expressed in nanoseconds. ... k' 18 a constant in appropriate units and [P'Tw'] is the average number of usable neutrons of energy E leaving the source per second. From equation 1 it is obvious that for a given resolution, R, and a given peak neutron production rate, P', the frequency should be high (w 1000 pulses per second) and the pulse wiatr small (to limits set later). 4. Intensity and Resolution Including Time Uncertainties Other Than from Pulsed Source The time resolution of a spectrometer 16 determined not only by the time width of the pulse of fast neutrons but also by the moderation time, the time jitter in the detector, the time uncertainty due to the time-of-flight in both the moderator and detector, and the analyzer channel wiath. They al. combine to give a total time resolution (added in quadrature) of Yw2 + (AT)2 where w 18 the width of the pulse of the fast neutrons and AT 18 the time resolution due to all other effects including the moderation time. Equation 1 must now be modified to R(2) = x(8) ** (7 +407) - (2) TE where P 18 the number of fast neutrons produced in the source peri second during the pulse and k(E) now includes the moderator efficiency (including the area factor) to moderate fast neutrons to usable neutrons of energy E. From equation 2 it is obvious that for a given pulsed source that if P and f are not dependent upon w, a given pulsed neutron source will be optimized if w is increased until it neerly equals AT. However, if the average number of neutrons produced in the source per second (Pfw] 18 constant, there is not reason to increase w even if it is smaller than AT. 5. Time Jitters of Detectors for Cross Section Measurements Let us consider the types of measurements we wish to make and the time jitters of the detectors appropriate for these measurements with present day detectors. For transmission measurements the most common detectors are BF, proportional counters, głº liquid scintillators, BPO- NaI scintillatɔrs, and Liº glass scintillators. The first two have time jitters of ~ 1000 and ~ 500 nanoseconds, respectively; but the latter two have time jitters of ~ 10 and 20 nanoseconds, respectively for 10% detection efficiencies, and this time jitter can be reduced with a lo88 in detector efficiency. For capture cross section measurements large liquid scintillators or a Moxon-Rae detector may be used. The former has a detection efficiency of ~90% and a time resolution of w25 nanoseconds, and the latter has an efficiency of 1% and a time resolution of ~ 3 nanoseconds. For f1881on measurements gas scintillators an with a time resolution do nanoseconds or solid state detectors with time resolutions ~ 20 nanosecondo may be used. For the measurement of gamma ray spectra with high gamma ray energy resolution, Li drifted germanium detectors' have time resolutions of ~ 20 nanoseconds. Thus, for all types of cro86 section measurements, detectors are available which have time jitters of ~ 20 nanoseconds or less. In the future one can hope for a factor of 2 or vre e more improvement. An excellent summary on detectors for the various types of cross section measurements has been made by Brooks. 6. Total Time Uncertainties Including Moderation Time Since several pulsed sources produce pulses which contain primarily fast neutrons, moderation 18 required to increase the intensity of neutrons with energies below ~ 10 keV. The time uncertainty produced for adequate soderation varies with neutron energy and 18 equal to 1400 nanoseconds. Since de ones. Since detectors Velin eV) can be made to have effective thicknesses 2 cm (to give a time jitter due to the flight time of the neutrons in the detector equal to the moderation 2000 nanoseconds. Hence, for 10 keV time), the total time uncertainty is is le (in eV) neutrons the total time uncertainty 1s 30 nanoseconds or less for suitable detectors for all types of measurements. By reducing the moderator and detector thicknesses (nence, decreasing the intensity of low energy neutrons and the efficiency of the detector), the time uncertainty can be reduced to < 20 nano- seconds or less. Whether this change will improve the quality of the data 18 DA -4* too difficult to predict and must be demonstrated experimentally. For measure- ments at higher energies (ww2.00 keV) time resolutions of 10 nanoseconds or less should be possible; for low energy measurements the time resolution 18 2000 nanoseconds 108. Therefore, the width of the burst of fast neutrons , Kishould YE(in eV) be kept less than these time uncertainties, AT, for the optimum puised neutron source. 7. Pulse Repetition Rates If these minimum burst widths are used, the flight paths and resolutions are such for all types of measurements that pulse repetition rates of ~ 1000 pulses per second can be used with the aid of glº filters or a mechnical shutter to reduce the overlap neutrons. For some lower resolution partial cross section measurements where high intensity 18 needed, higher repetition 800 er rates could be utilized. Naturally if pulse widths are wider than the optimum values (1.e., >AT), somewhat lower pulse repetition rates must be used. 8. Parameters of Several pulsed Sources and Approximate Comparison To compare various pulsed neutron sources using equntion 2 18 difficult because of the uncertainty in the factor k(E). This factor includes the moderator efficiency to moderate the fast neutrons to lower energies and the fraction of these lover energy neutrons which can be used due to the spatial extent of the moderator. For example, if only a small sized sample were available for a transmission measurement, the usuable fraction of the neutrons from the moderator 18 smaller (given approximately by the ra:10 of the sample area to the moderator area). For partial cross section reasurements or transmission measurements with large samples, the fraction 18 much higher and limited only by the size of the beam tube relative to the size of the moderator. The energy dependence of this factor k(E) differs for the various pulseü neutron sources, and a comparison ignoring this factor cannot be expected to be meaningful, except to one or two orders of magnitude. Table I lists some of the specifications for several existing neutron sources used for neutron spectroscopy. For the fast chopper at the ORR, P was taken to be 3 x 2074 since the effective area of the moderator which is used is only 2 cm?. In Table I, AT was assumed to be equal to 102 + (2000/4Elev))?)', except for the chopper. From this crude comparison, it is apparent that the pulsed accelerator sources and à nuclear explosion are 3 to 7 orders of magnitude (10% - 10%) more intense at 100 keV than the USSR pulsed reactor or the ORR chopper; but the ratio drops for lower neutron energies. The booster on the Harwell linac with wide pulses are: is of no advantage for measurements at high energies. At low energies a factor of 10 gain is due to the booster; the other factor of ~ 10 18 due to the increased burst width from the accelerator. A more detailed discussion of this comparison was made two years ago by Rae for various energy resolutions. The results are shown in slides 1 and 2. The curve labelled H-10 was the Herwell linac two years ago with 10 nanosecond bursts, and H-10ES was the estimated future performance which is now realized. The curves labelled HB-250 and HB-1000 are for the Harwell KUIV2: booster with 250 and 1000 nanosecond bursts from the accelerator. The curves labelled C-120 and C-20 are for the Nevis spectrometer with 120 and 20 nano- second pulses. The two numbers in parentheses are the repetition rates and flight paths which were assumed for the comparison. Slide 3 shows the results of a comparison of various sources made by J. A. Harvey and F. C. Malenschein' to show the superiority of the proposed ORNL : linac. The comparisons were made for resolutions equal to Doppler widths below 10 keV and A E/E = 5 x 20-4 above 10 keV. Estimates were made of the moderator efficiency, the availability of the accelerator for pulsed neutron experiments, the number of flight paths used simultaneously, etc. to obtain the over-all figures of merit. Table II shows the details of the comparison at 10 keV. Similar studies have been made by E. B. Paul“, R. G. Fluhartyand A. B. Smith with the conclusion that the pulsed accelerators are far superior to choppers at high energies. Fluharty concluded that at eV energies the addition of a booster to a pulsed accelerator might result in gains as high as 80 over a similar accelerator with no booster. To make a more detailed comparison of the various pulsed sources ces meaningful, it is necessary to examine the experimental data which the various spectrometers have produced in the past few years. However, it 18 still difficult to make an accurate comparison because different detectors were used. . We shall assume that the detector which was used was selected because it gave the best results, for that particular spectrometer. 9. Comparison of Spectrometers for High Resolution Measurements Above ~2 keV Let us first consider transmission measurements which have been made with the highest resolution possible for each spectrometer. Since I 3.86 ume we are least familiar with the results from the USSR pulsed reactor spectrometer, I will show a few slides of their work. slide 4 shows their experimental arrangement with several flight stations. Transmission measurements have been made with the 1000 meter flight path with a resolution of 40 nanoseconds/meter using a 2000 cm? area detector with an efficiency of . 50% at 100 ev.' The resolution obtained is shown in slide 5 (taken from a paper by Zabiyakin). Slide 6 shows some data obtained on rhodium”; above ~300 eV the resolution is not sufficient to observe all the resonancee. Several fast chopper spectrometers in the U.S. have produced data with resolutions of ~ 20 nanoseconds/meter. Since the intensity varies as the third power of the resolution, fast choppers at high flux steady state reactors are about an order of magnitude better than the IBR at 3 kW. Also, only a few cm of samples are needed for transmission measurements with the chopper vs several hundred cm? for the IBR which is another factor of 100 in favor of seve choppers if samples are available only in small quantities. Transmission measurements upon rhodium have also been made using the Saclay electron linac" with resolutions as small as 2.5 nanoseconds per meter at the higher energies. Recently, measurements have been made at Saclay with resolutions of 1 nanosecond per meter. Measuriments with the Harwell linac and the. booster have achieved a resolution of 1.6 nanoseconds/meter. - Finally transmission measurements have been made with the NEVIS spectrometer with & resolutions as low as 0.5 nanoseconds per meter at the higher energies. It. is doubtful that the pulsed reactor spectrometer, even if operated at 30 MW average power, could ever produce data with this high resolution (and flight paths would .. impossibly long, 10 meters). Even with a fast chopper on a pulsed reactor operating at 30 MW, I don't believe it would ever produce the high resolution of 0.5 nanoseconds/meter now available above ~ 2 kev. 10. Comparison of Spectrometers for the Energy Region ~ 200 to w 2000 eV For partial cross section measurements, resolution must be sacrificed for intensity; and we shall now compare various spectrometers for these somewhat 10 12 lower resolutions. Slide 7 shows capture cross section data obtained using the IBR spectrometer with two 200 liter liquid scintillators at the 750 meter flight station resulting in a resolution of 55 nenosec:onda/meter. “2 Capture measurements have been made at the RPI linac with a 1200 liter scintillator at 25. meters with a resolution ~7 nanoseconds/meter above a few hundred ev. slide 8 shows data obtained from a Moxon-Rae capture detector at 90 meters with the Aarwell linac and the booster resulting in a resolution - 3 nano- seconds/meter, Fission measurements' have been made with the IBR spectrometer using . a 400 liter liquid scintillator containing cadmium : propionate at, 1000 meters. ... im.. ' Fission is recorded by delayed coincidences between the pulse associated with the instantaneous gamma rays and the pulse originating by the slowed-down neutron by capture in cadmium (Diven's technique). This technique is possible by virtue of the long 40 usec pulses of the pulsed reactor. The same detector is also used to measure the capture cross sections of the fissionable_nucleſ. Slide 9 shows results from u235. Fission and capture measurements upon us have been made on the .. RPI linac by G. deSaussure, et al. with a resolution ~ 5 nanoseconds/meter above a few hundred electron volts. Fission cross section measurements have been made at NEVIS and with the electron accelerators at Saclay, Harwell, and Livermore with resolutions of w 10 nanoseconds/meter. A recent nuclear explosion exper:iment by Diven et al. has produced fission cross section data above I keV energy with a resolvtion of 2 nanoseconds/meter. 2 . 11 For scattering cross section measurements at IBR. (made to determine mee menn the spin states of resonances),a scintillation detector containing several alternate layers of (ZnS + Blº) and plexiglass was located at the 500 meter flight station. The neutron lifetime in this detector is ~ 15 usec. Scattering measurements have been made at Harwell with a lithium-loaded glass scintillator at a 50 meter flight path and a resolution ; 10 times better than the man resolution used at IBR....... Therefore, for all the partial cross section measurements which can presently be made with resolutions - 5 nanoseconds/meter above ~ 200 eV, it does not appear that the pulsed reactor (with or without a fast chopper) would be a serious competitor unless average power levels of several tens of mega- watts with a duty cycle of 1 part in ~ 1000 could be obtained. 11. Comparison of Spectrometers in the Electron Volt Energy Region For partial cross section measurements at lower neutron energies the pulsed reactor competes more favorably with the pulsed accelerators. However, 1t 18 difficult to find appropriate data to make a meaningful comparison. In the electron volt energy region a pulsed reactor operating with an average power of a few hundred kilowatts with a duty cycle of 1 part in ~ 100 would probably be comparable to a pulsed accelerator spectrometer, but a pulsed accelerator with a booster would likely be one to two orders of magnitude · superior. A similar conclusion was obtained by Bunin' for the combination of the IBR and an electron cyclotron. Using 1 microsecond pulses from this cyclotron, which produces 15 x 105 neutrons/ second in the pulse and with a multiplication factor of 80 from the IBR, 4 x 1011 neutrons win! - - - . - - - - --- - - -- - 12 per pulse are produced. At a repetition rate of 83 pulses per second the average reactor power 16 600 watts. Bunin concludes that this arrangement 1.8 165 times better than the pulsed reactor alone operating at an average power of 3 KW. If the comparison were made with RPI linac which can producem50 times more neutronsc :: per pulse than the electron cyclotron used on the IBR, the pulsed reactor would have to operate at w 10 megawatts of average power, with peak. powers of thousands of megawatts to equal the pulsed-accelerator pulsed-multiplier, combination. 12. Ultimate Limit for a Pulsed Spectrometer · In the comparison of these pulsed sources no consideration has been given to the limitation imposed upon the pulsed accelerator sources due to heat removal. At present, an electron linac is capable of producing a bean of only v 50 KW average power; and a proton machine such as Nevis produces only a few hundred watts of heat. If a booster with a neutron gain of 100 is added to a 50 KW electron accelerator, it will still produce only ~ 500 KW of heat in a volume of several liters. (A 100 MeV electron beam requires ~ 2000 MeV per neutron, and only 200 MeV per neutron is needed for in fission.) When heat reraoval becomes the limiting factor for a pulsed fission neutron source, a factor of ~ 5 can be gained by the use of a w 800 MeV proton accelerator as proposed by Chalk River, since only 40 MeV is required per neutron. 13. Backgrounds For most of the cross section measurements referred to in this paper, backgrounds have been of some importance - some times as high as 50%. With a fast chopper spectrometer, backgrounds are bothersome in some experiments and a factor of 20 or 100 reduction by use of a pulsed reactor would certainly be welcomed. The background from the delayed neutrons in the booster at Harwell 23 is definitely a disadvantage for measurements at low neutron energies with a black detector. Of course, a mechanical shutter (as has been installed on the NEVIS cyclotron) would reduce this background. Instead of a booster, a pulsed multiplier could be used which has a lower multiplication between bursts than during the burst from the accelerator. Fluharty has also proposed circulating the fuel to remove the fission products. In general, the signal to background ratio varies inversely as the pulse repetition rate for similar pulsed sources with the same average neutron production rate. This point has been emphasized by Smith in considering the use of the zgs as a spectrometer and is obviously carried to its limit with a nuclear explosion. 14. Conclusions (1) For high resolution measurement above ~ 2 keV where resolutions 0.5 nanoseconds/meter or better are often required, existing pulsed accelerators are far superior to a pulsed reactor. (2) For measurements from ~ 200 to ~ 2000 eV where resolutions of N 5 nanoseconds/meter are sometimes required, a pulsed accelerator with a TO pulsed multiplier having a neutron multiplication ~10 is somewhat superior to the pulsed accelerator alone and is estimated to be approximately equivalent to a pulsed reactor-chopper combination at an average power of tens of mega- watts and a peak power of tens of thousands of megawatts. (3) For low-resolution, high-intensity measurements in the electron volt energy region, a pulsed accelerator with a pulsed multiplier having a neutron multiplication of ~ 100 would be approximately equivalent to a pulsed reactor with an average power of a few megawatts and a peak power of hundreds of megawatts. References 1. F. D. Brooks, Neutron Time-of-Flight Conference, Saclay 1961, page 389. 2. E. R. Rae, corrigenda (February 1963) to Pulsed Accelerator Time-of-Flight Spectrometers, AERE-NP/GEN 21 issued March 1962. 3. J. A. Harvey and F. C. Maienschein, Comparison of Pulsed Neutron Sources, ORNL-IM-582, May, 1963. 4. E. B. Paul, Neutron Time-of-Flight Conference, Saclay 1961, page 375. 5. R. G. Fluharty, Neutron Time-of-Flight Conference, Saclay 1961, page 383. 6. Zabiyakin, G. I. et al., page 42 AEC-tr-5734, Proceedings of the Working Conference on Slow Neutron Physics (1961). 7. Bunin, B. N., et al., Third International Conf. on Peaceful Uses of Atomic Energy, Geneva 1964, Paper 28/P/324. 8. Wang Nai-yen, et al., page 48 AEC-tr-5734, Proceedings of the Working Conference on Slow Neutron Physics (1961). 9. Ribon, P., et al., Journal de Physique et Rad. 22, 708 (October 1961). See also Ribon, P., et al., Neutron Time-of-Flight Conference, Saclay 1961, page 97. 10. C. A. Uttley and R. H. Jones, Neutron Time-of-Flight Conference, Saclay 1961, page 109. 11. J. B. Garg, J. Rainwater and W. W. Havens, Jr., Phys. Rev. 131, B547 (1965); see also Phys. Rev. 136, B277.(1964); and Phys. Rev. 134, B985 (1964). 12. Kim H1 Sang, et al., page 52, AEC-tr-5734, Proceedings of the Working Conference on Slow Neutron Physics (1961). TABLE I Approximate.. Comparison of Pulsed Neutron Sources Neutron Source (neutrons/sec) (sec-?) .. (n sec) | 105 ev 1x1024 +(AT)2 103 ev 1x2014 IBR (1) 1.3x2018 3.3 10 eV 1x1024 3x1024 (pulsed reactor) 3x2024 1000 3x1024 3x1024 Fast chopper, (steady state reactor) Electron linac, Harwell (3) 3x1026 no booster actor) 8x2027 4x2026 5x1024 Electron linac, Harw with booster | 7x1027 7x1027 1x1077 NEVIS(4) 1.1.x2019 2x1029 7x2020 4x2028 4x2016 5X2020 1x1019 Nuclear explosion (5) ~ 2030 6x20-8 (1) Reference 6. Reference 3. (3) Reference 2. (4) J. R. Rainwater, Neutron Time-of-Flight Conference, Saclay 1961, page 321; also Rev. Sci. Instr. 35, 263 (1964). (5) G. A. Cowan, Neutron Time-of-Flight Conference, Saclay 1961, page 367. . + Tabia ila Figures of klorit for oʻar Pulaod Neutron Sources 8€.se 3x10 apak-28* 16 • X1,-1.7 Lene Dame Finive taget med Monerator Vrable Time Spread Die der Video a Terget Thickness, Moderacion Time, Droctor Thickorn Electronic Flipke Pada lor Eero Larolucion Equal to Demples broadenios for A - 100 Wasia. Vaade Lapraidios hace Combined Maria Teget Moderator plus Detector midors wa Med Fiske Pads Uuod Sacalassonely Availability for and Nerede Experienta Marie for One Overall Figure precise of Merit lor Equals Poeilis. Concias 1. IND KEJAEY ST Cerc) (sec" 0. 1,000 X,000 4 1,000 3.000 20 0.500) (190.000) <, Rh 0:17. stehenden en n rai . 14. 2. Number of levels of rhodium u neutron emery. i functim of the 1.** in SLIDE 6 boten. -- -.. . -.-.-- . ... . v wallad . r ORNI - AEC - OFFICIAL im. is the income intorno Fig. 3. Tino spectrum of rhodium, obtained using a captured game-radiation detector. min.............. L . SLIDE 7 non cura cous con olenna . - ...--- iny UUWI ONDOR LUI IMINUT: SLIDE 9 SLIDE 8 OOO OOK Wow www me ----- .. az. 10. nno C20 O ULTIMWhDMIN MINI ORNL - AEC - OFFICIAL OIX! - AEC - OFFICIAL END .: . . AXT DATE FILMED 8 / 17 / 65) 4 .. . 2. l '