ORNL-TM=-3151 Contract No. W-ThO5-eng-26 Reactor Division A STUDY OF FISSION PRODUCTS IN THE MOLTEN-SALT REACTOR EXPERIMENT BY GAMMA SPECTROMETRY A. Houtzeel F. F. Dyer AUGUST 1972 NOTICE This document contains information of a preliminary nature and was prepared primarily for internal use at the Cak Ridge National Laboratory. It is subject to revision or correction and therefore does not represent a final report. NOTICE is report was prepared as an account of work sponsared by the United States Government, Neither the United States nor the United States Atomic Energy Commission, nor any of their employees, nor any of their contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, com- pleteness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. OAK RIDGE NATIONAL LABORATORY Oak Ridge, Tennessee 37830 operated by UNION CARBIDE CORPORATION for the U.S. ATOMIC ENERGY COMMISSION S ey 15 URLIME plsTRIBUTION or TWIS BOCUNENT 158 %fl iii CONTENTS ACKNOWLEDGMENT « « v ¢ & o & 4 ¢« o & o o & SUMMARY l. . . - . . . . . - » » - 4 » - . INTRODUCTION TO THE MOLTEN SALT REACTOR EXPERIMENT. 1.1 1.2 Molten-Salt Reactor Concept . . . Description of the MSRE . . . . . OBJECTIVES OF THIS GAMMA~RAY SPECTROMETRY DESCRIPTION AND PERFORMANCE OF EQUIPMENT. 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 Background . . . . .+ . ¢ ¢ . . . General Description . . « . . . . Detector and Amplifier . . . . . Analvzer . + + ¢« o o s e s o0 e e Collimator Assembly . . . . .« . . Setup of Experimental Equipment . Locational Equipment . . . . . . Shielding e h e e e e e e s Calibration Setup . « « « o« ¢ &+ & ANALYSIS OF SPECTRA « « ¢« ¢ 4+ ¢« & ¢« o & 4b.l 4.2 4.3 4.4 Purpose and General Procedure . . Problems . + ¢« ¢+ ¢« ¢ o o « ¢ o« & Table of Isctopes . . + « « « = Computer Programs . . + « « « + o CALIBRATION . o 4 & ¢ o o o o o o o = 5.1 5.2 5.3 5.4 5.5 5.6 5.7 General . . . « ¢ 4« o o 4 e 4 e Source .+ . ¢ o e 4 s e s e e s s Slit Experiment — Source Strength Single Source Experiments . . . . Heat Exchanger Calibration . . . Calibration of Shielding Materials - Calibration for Fission Gases in the MSRE MEASUREMENTS ., , . . « « ¢« v ¢ o« o o & Page vii 11 13 13 14 17 17 18 20 21 21 24 27 27 28 30 31 33 33 34 35 43 b 48 52 53 . o . - RESULTS . « ¢« & v ¢« o « + + . 7.1 Group A Spectra . . . . . 7.1.1 Heat exchanger . . 7.1.2 Main reactor off-gas line 7.1.3 Main fuel lines . . . 7.2 Group B Spectra . . . . 7.2.1 Main reactor off-gas line 7.2.2 Beat exchanger . . . . 7.3 Group D Spectra . . . . 7.4 Group E Spectra . . . e e 7.5 Group F Spectra . 7.5.1 Main reactor off-gas line 7.5.2 Heat exchanger . 7.5.3 Drain tank . . . 7.6 Group G Spectra . . . . 7.6.1 Heat exchanger . 7.6.2 Main reactor off-gas line . 7.6.3 Fuel pump bowl and fuel 7.7 Group 4 Svoectra . 7.7.1 Heat exchanger . . 7.8 Group I Spectra . . . 7.9 Group J Spectra . . . . . 7.10 Group K Spectra . 7.11 Group L Spectra . e e e 7.12 Group M Spectra . . . . CONCLUSTIONS o . « e e . 8.1 Data Collection and Analysis 8.1.1 Gamma spectrometer system , 8.1.2 Calibration . . . 8.1.3 Computer analysis program 8.2 Results — General . o . lines 8.2.1 Metal surfaces in direct contact with fuel salt . 8.2.2 Main reactor off-gas line . 134 136 136 138 141 141 141 142 143 143 143 145 8.3 Elements and Nuclide Chains . . . . 8.3.1 Niobium . . « « « « « « o« + . 8.3.2 Ruthenium-rhodium . . . . . . 8.3.3 Antimony-tellurium~iodine . . 8.3.4 Extrapolation back to reactor 9, EPILOGUE . o &+ ¢ &+ ¢ o o o s o s o s o o & Appendix A. TABLE OF RADIONUCLIDES . . . . . Appendix B. PRODUCTION AND STANDARDIZATION OF shutdown time . 11OMAs IN SILVER TUBING FOR MSRE GAMMA SPECTROMETRY . . . . . Appendix C. ABSOLUTE EFFICIENCY CURVES OF THE GAMMA-RAY DETECTION SYSTEM USED FOR THE CALCULATION OF ABSOLUTE AMOUNTS OF NUCLIDES DEPOSITED . . . . . Appendix D. PRINCIPLES . & & v &+ & o & + & CALCULATIONS OF COUNTING EFFICIENCIES FROM FIRST Page 146 146 148 151 154 155 161 181 188 194 vii ACKNOWLEDGMENT The authors gratefully acknowledge the contributions of the many per- sons who were instrumental in the proper execution of this experiment, es-— pecially in the design of the equipment, the recording of the spectra, and the analysis of the results. In particular, we appreciate the valiant help of the following persons, R. Blumberg, for the very helpful efforts in the design of the equip- ment and the recording of the spectra. J. R. Engel, for his wise advice during the course of the experiment. A, F, Joseph, for his effort in making the spectrum analysis program operable on the ORNL computers., J. L. Rutherford, for the help in analyzing the computer results and preparing the figures. L. P. Pugh, for the assistance in the design of the equipment, J. A. Watts, for his help in analyzing the computer results and preparing the figures. SUMMARY The operation of the Molten-Salt Reactor Experiment (MSRE) has demon- strated that a mixture of fluoride salts is a practical fluid fuel that is quite stable under reactor conditions. The chemistry of the fission prod- ucts is such, however, that some of them leave the circulating fuel salt and appear on the moderator graphite‘in the core, on the metal surfaces ex- posed to the salt, and on the metal surfaces in the off-gas system. Xenon and krypton fission gases are stripped in the off-gas system, where they decay to daughter nuclides. Some other elements (Mo, Nb, Ru, Te, and Sb) appear to exist in the metallic state and tend to plate out on metal or graphite surfaces or be carried out into the off-gas system as particles. Because it is important in the design of larger molten-salt reactor systems to know where and in what proportion fission products are distributed throughout the system, considerable efforts were made to obtain this infor- mation in the MSRE. A technique was developed at ORNL to locate and measure fission prod- uct depositions on surfaces exposed to the salt and in the off-gas system of the MSRE by the intensity and energy spectrum of the emitted gamma rays. A gamma-ray spectrometer was developed, consisting of a Ge(Li) detector, a 4096-channel analyzer, and a lead collimator to permit examination of small areas., This device was usually positioned with the MSRE portable mainte- nance shield over the different reactor system components, with precise alignment and location achieved by a laser beam and surveyor's transits. Measurements were made not only with the reactor system shut down and drained but also with the fuel circulating and the reactor at several power levels, Altogether some 1000 spectra were taken, 257 of which were recorded with the reactor at some power level (a few watts to full reactor power). Another 400 spectra were taken to calibrate the equipment. Com- puterized data handling permitted this mass of data to be analyzed quali- tatively and quantitatively. Most of the effort was focused on the off-gas system and the primary heat exchanger, the latter because it contains 407 of the metal surfaces exposed to the salt. The off-gas system contained not only fission prod- ucts with gaseous precursors but also metallic elements with their decay products [such as Nb, Mo, Ru, Sb, Te(I)]. The MSRE heat exchanger con- tained mostly depositions of the same metallic elements; It was observed that fission gases form a major source of activity in the heat exchanger when the reactor is shut down and the fuel is drained immediately (emer- gency drain). A high~resolution gamma-ray spectrometer used with proper remote main- tenance equipment and location tools proved to be very vefsatile in locating and evaluating fission product depositions in a highly radioactive reactor system. 1. INTRODUCTION TO THE MOLTEN-SALT REACTOR EXPERIMENT 1.1 Molten-Salt Reactor Concept’:? The molten-salt reactor concept originated in 1947 at Oak Ridge as a system for jet aircraft propulsion. The idea was to use a molten mixture of fluoride salts including UF, as a fuel that could be circulated to re- move the heat from the core. Fluoride salts looked promising because of their basic physical chemistry: the vapor pressure of the molten salts would be extremely low and they would not react violently on exposure to air or water. Radiation damage would be nonexistent in the completely ionic liquid fluorides. Since a variety of interesting fluorides (NaF, LiF, BeF., ZrF,, UF,, etc.) were known to be stable in contact with some common structural metals, a corrosionless system seemed attainable. 1In addition, high specific heat, good thermal conductivity, and reasonable viscosity made these liquids good heat transfer media. On the other hand, the high melting point of potential fuel mixtures (400—500°C), while no drawback during power operation, would require provisions for preheating the piping and keeping the salt molten during shutdowns. An intensive ef- fort on molten salt was undertaken at the Oak Ridge National Laboratory, and by 1954 a molten-salt reactor was operating at temperatures around 800°C, It had been recognized that the technology of the molten~-salt system was well suited for the development of a commercial Th-22°U power reactor. Thus, in 1956 the Molten-Salt Reactor Program was established at ORNL, By 1960 a picture of an economically attractive molten-salt reactor had come into focus. Its core would contain the graphite moderator in di- rect contact with molten salt flowing through channels and there would be either one or two salt streams. If one, it would contain both thorium and uranium, giving a high-performance converter or even a breeder with a small breeding ratio. 'P. N. Haubenreich, '"Molten-Salt Reactor Progress," Nucl. Eng, Int, 14(155), 325-29 (April 1969). M. W. Rosenthal, P. R. Kasten, and R. B. Briggs, '"Molten-Salt Reac- tors — History, Status and Potential," Nucl. Appl. Tech. 8(2), 107-17 (February 1970). The basic technical feasibility of the molten-salt reactors was on a sound footing — a compatible combination of salt, graphite, and container metal. A salt mixture based on 'LiF and BeF, looked most attractive from the standpoint of melting point, viscosity, neutron absorption, and freedom from mass transfer. A nickel-base alloy, INOR-8, had been developed that was practically unaffected by the salt at temperatures to 700°C, that was superior in strength to austenitic stainless steel, and that was susceptible to conventional fabrication. It was found that salt did not wet or react significantly with graphite and that, by reducing the graphite pore size, intrusion of salt into the graphite could be prevented. Although the ma- terial situation was encouraging and test loops had operated successfully, a reactor experiment was needed to really prove the technology. Therefore the objective of the Molten-Salt Reactor Experiment (MSRE) was to demon- strate that the key features of the proposed breeders could be operated safely and reliably and maintained without excessive difficulty. 1.2 Description of the MSRE?® The MSRE was to use essentially the same materials as the breeders. There was no attempt to design it to be a breeder, however, since this would have entailed added expense and complexity in the form of a large core or a blanket of fertile material. Some of the important design cri- teria were: . core of bare graphite with fuel flowing in channels, . removable specimens of graphite and metal in the core, . provision for sampling the salt and adding uranium during operation, . power 10 MW or less, 1 2 3 4, fuel temperature around 650°C, 5 6 heat rejected to the air via a secondary salt loop, 7 . fuel pump rather larger than necessary (to minimize scaleup to the next reactor), 3p. N. Haubenreich and J. R. Engel, "Experience with the MSRE," Nucl. Appl. Tech, 8(2), 118—36 (February 1970). 8. simplicity and conservatism to enhance reliability, 9. zero leakage of salt in operation, 10. enclosure capable of safely containing spill of entire fuel. The flowsheet that was arrived at is shown as Fig. 1.1. Details of the MSRE core and reactor vessel are shown in Fig. 1.2. The 55-in.~diam core was made up of graphite bars, 2 in. square and 64 in. long, with flow passages machined into the faces of the bars. The graphite was especially produced to limit pore size to 4 u to keep out the salt, All metal components in contact with molten salt were made of Hastelloy N (a commercial version of INOR-8), which had been approved for construction under ASME codes., The three control rods were flexible, consisting of hollow cylinders of Gd;03-Al.0; ceramic canned in Inconel and threaded on a stainless steel hose. Draining the fuel provided positive and complete shutdown. The volute of the centrifugal fuel pump was enclosed in a tank (the pump bowl) which was the high point in the loop. The pump suction was open to the salt in the bowl, so that the pump bowl and the connected overflow tank provided the surge space for the loop. A blanket of helium, generally at 5 psig, was provided over the salt. A tube into the top of the pump bowl connected to the sampler-enricher, which was a two-chambered, shielded transfer box; small sample buckets or capsules containing uranium~rich salt could be lowered from this transfer box into the pool in the pump bowl. A spray ring in the top of the fuel pump bowl took about 4% of the pump dis- charge and sprayed it through the gas above the salt to provide contact between helium and fuel salt so that the gaseous fission products could escape into the gas. A flow of 4 liters/min (STP) of helium carried, among others, the fission gases such as xenon and krypton out of the pump bowl, through a holdup volume, a filter station, and a pressure~control valve to the charcoal beds. The beds operated on a continuous-flow basis to delay xenon for about 90 days and krypton for about 7 1/2 days, so only stable or long-~lived nuclides could get through. All salt piping and vessels were electrically heated to prepare for salt filling and to keep the salt molten when there was no nuclear power. The air-cooled radiator was equipped with doors that dropped to block the | . o oo . . ' . . ORNL-DWG 6% -{1410R : : ‘ S psig 5 psig FUEL i oot PUMP t SAMPLER- o i . LEGEND i JJenricher 1 JISAMPLER s FUEL SALT - -—— —r i i om— COOLANT SALT ; I ; : To ABSG'U?E FlLTERS""""j " TsasEnsaanIAN %Lw COVER GAS ) - %_D ! 015 %F i ———=—= RADIOACTIVE OFF -GAS _ Loy Loner 850 G.PM. OFF-GAS | ] ' HOL DUP HEAT EXCHANGER [* 1210°F ey OVERFLOW TANK - 11 | . ABSOLUTE 70 °F AR FLOW: 200,000 cfm FILTERS 1200 GFM, v T ——————r . BLDG REACTOR 1075°F " VENTILATION ; VESSEL POWER FREEZE FLANGE (TYF) —— el ‘ ‘ A ‘ oM STACK | FAN l T 1 ‘ v P .o FROM i . sl COOLANT FREEZE VALVE (TYP) . s ;‘:1 SYSTEM ;_l . 2 RADIATOR L f 4o i‘” = i TEmTmTTTT $ UTTTTO lE ‘g L L & _ . i £ 2 : 10 | co Wy e T ¥ : : l : . ‘ m—y . . : 1!_0 D ABSOLUTE t _ b p o ! WATER STEAM 4. P 2-_?_ *-E_f_b il FILTERS { ‘ WATER STEAM . ; i i { . S . . \ ' Y i : s wan i : i U v ey [ geelansd CHARCOAL, 'E : A T eED . i . i ! COOLANT . DRAIN TANK ' ( N/ | : e romen e e st i o e e i e s Si FLUORIDE BED Fig. 1.1. Design flowsheet for the MSRE. ORNL-LR-DWG 61097RtA FLEXIBLE CONDUIT TO CONTROL RCD DRIVES GRAPHITE SAMPLE ACCESS PORT oy 4 /& COOLING AIR LINES ACCESS PORT COOLING JACKETS REACTCR ACCESS PORT SMALL GRAPHITE SAMPLES HOLD -DOWN ROD OUTLET STRAINER CORE ROD THIMBLES LARGE GRAPHITE SAMPLES - CORE CENTERING GRID FLOW DISTRIBUTOR VOLUTE GRAPHITE - MODERATOR ¥ STRINGER FUEL INLET CORE WALL COOLING ANNULUS REACTOR CORE CAN | REACTOR VESSEL — T ANTI-SWIRL VANES MODERATOR VESSEL DRAIN LINE SUPPORT GRID Fig. 1.2. Details of the MSRE core and vessel. air duct and seal the radiator enclosure if the coolant-salt circulation stopped and there was danger of freezing salt in the tubes. There were no mechanical valves in the salt piping; instead, flow was blocked by plugs of salt frozen in flattened sections of certain auxiliary lines. Tempera=- tures in the freeze valves in the fuel and coolant drain lines were con- trolled so they would thaw in 10 to 15 min when a drain was requested. The drain tanks were almost as large as the reactor vessel, but the molten fuel was safely subcritical because it was undermoderated. Water-cooled bayonet tubes extended down into thimbles in the drain tanks to remove up to 100 kW of decay heat if necessary. The physical arrangement of the equipment is shown in Fig. 1.3. The reactor and drain tank cells are connected by a large duct, so they form a single containment vessel. The tops of the two cells consist of two layers of concrete blocks, with a weld-sealed stainless steel sheet between the layers; the top layer .is fastened down. The reactor and drain tank cell were kept at -2 psig during operation. A small bleed of nitrogen into the cell kept the oxygen content at- 3% to preclude fire if fuel pump lubri- cating oil should spill on hot surfaces. A water-cooled shield around the reactor vessel absorbed most of the escaping neutron and gamma-~ray energy. The 5-in. salt piping in the reactor cell included flanges that would per- mit removal of the fuel pump or the heat exchanger. The flanges were made unusually large and were left uninsulated so that salt would freeze be- tween the faces. All the components in the reactor and drain tank cells were designed and laid out so they could be removed by the use of long-handled tools from above. When maintenance was to be done, the fuel was secured in a drain tank and the connecting lines frozen. The upper layer of blocks was re- moved and a hole cut in the membrane over the item to be worked on; after a steel work shield (the portable maintenance shield) consisting of two parts was set in place, a lower block was removed. Then the two parts of this portable maintenance shield were moved together, and the hole, caused by the removal of the lower block, was covered. Through 5-in.~diam holes in the portable maintenance shield, one could then work remotely in the re- actor cell. Fig. 1.3. REMOTE MAINTENANCE CONTROL ROOM - 1. REACTOR VESSEL . HEAT EXCHANGER FUEL PUMP FREEZE FLANGE . THERMAL. SHIELD . COOLANT PUMP O AHGN Layout of the MSRE. ORNL-DWG 63-1209R RADIATCR . COOLANT DRAIN TANK . FANS - . FUEL DRAIN TANKS . FLUSH TANK . CONTAINMENT VESSEL . FREEZE VALVE 10 The conventional instrumentation and control systems for the reactor were augmented by a digital computer that was used to log data and help analyze the operation. About 280 analog signals from the reactor were wired to the computer. Construction of the primary system components for the MSRE started in 1962, and installation of the salt systems was completed in mid-1964. Pre- nuclear tests, in which first flush salt and then fuel carrier salt con- taining no uranium were circulated more than 1000 hr, showed that all sys- 2337 was then added to the carrier salt as the tems worked well. Enriched UF,-LiF eutectic (61 wt % U) and on June 1, 1965, criticality was achieved. In May 1966, the full power of 7.3 MW was reached. The shutdown in March 1968 was the end of nuclear operation with *3°U. Sufficient 22U had become available, and plans had been made to substitute it for the *2°U in the MSRE fuel to measure directly some nuclear charac-— teristics of great importance to the nolten-salt breeder design. After shakedown of the processing equipment, the flush salt and the fuel salt were fluorinated to recover the 218 kg of uranium in them. Uranium-233 was then loaded into the stripped fuel carrier salt, and criticality was attained in October 1968, after the addition of 33 kg of uranium (917 233 ). Nuclear operation with *°°U fuel continued until December 1969, when after 4 1/2 vears of successful operation, the reactor was shut down for the last . 4 time. “M. W. Rosenthal et al., "Recent Progress in Molten-Salt Reactor Development,'" TAEA At. Energ. Rev. 9(3), 60150 (September 1971). 11 2, OBJECTIVES OF THIS GAMMA-RAY SPECTROMETRY STUDY The operation of the MSRE demonstrated that a mixture of fluoride salts is stable under reactor conditions and that the majority of the fis- sion products remain with the circulating fuel salt; however, some fission products are found on the moderator graphite in the core, on the metal sur- faces exposed to the salt, and in the reactor off-gas system. TFor example, some elements (Mo, Nb, Ru, Te, and Sb) appear to exist in the metallic state and tend to plate out on surfaces in contact with the salt or to be carried into the off-gas system as particles. The behavior of certain fission products, especially those that vola- tilize or deposit, is of interest for several reasons in a molten-salt system. 1. The reactor chemists, of course, seek to understand the chemistry of the fission products in the salt. 2. The shielding required in remote maintenance of reactor components is strongly influenced by the total amount of highly active fission prod- ucts deposited in those components. 3. The deposited fission products may represent several megawatts of de- cay heat, creating a cooling problem after reactor shutdown and drain in a large, high-power molten-salt reactor. 4, Fission products that concentrate in the core by deposition on graphite would absorb more neutrons and hence reduce the breeding performance of a molten-salt reactor. For these reasons a comprehensive program of studies of fission product behavior in the MSRE was undertaken. The objective of the study described in this report was to determine the identity and magnitude of radiocactive fission product deposits in certain MSRE components using the technique of remote gamma-ray spectrometry. Particular attention was directed to the reactor off-gas system and the heat exchanger, the latter because it con- tained approximately 407 of the metal surfaces exposed to the circulating fuel salt. The deposition on graphite and metal in the core and in the 12 pump bowl is being studied by others and is discussed in other reports., > This report presents the results of the remote gamma-ray spectrometry in a readily usable form with some interpretations that may be useful in the overall effort of understanding the behavior of fission products in the MSRE. ®F. F. Blankenship et al., MSR Program Semiannu. Progr. Rep. Aug. 31, 1969, ORNL-4449, pp. 104—9. °C. H. Gabbard, MSE Program Semiannu. Progr. Rep. Feb. 28, 1970, ORNL-4548, p. 13. ’F. F. Blankenship et al., ibid., pp. 104-8. °F. F. Blankenship et al., MSR Program Semiannu. Progr. Rep. Aug. 31, 1970, ORNL-4622, pp. 60—70. 13 3. DESCRIPTION AND PERFORMANCE OF EQUIPMENT 3.1 Background The equipment that was used to obtain the data in this report (de- scribed below) was developed over a two-year period. In 1967 Blumberg, Mauney, and Scott? began to study devices for lo- cating and evaluating amounts of radicactive materials in high-radiation- background areas. During a shutdown of the MSRE in May 1967, they mapped the intensity of radiation coming from the fuel heat exchanger using a gamma-ray dosimeter mounted over a collimator in the portable maintenance shield. During the same shutdown a few data on energy spectra were obtained with a sodium iodide crystal mounted in a lead shield with a collimating 235 hole. At the end of U operation in March 1968, they made more, better measurements of gamma spectra by using a different collimator-shield com- bination, a lithium-drifted germanium diode, and a 400-channel analyzer.'® The conclusion then was that remote determination of fission product depo- gition by gamma spectrometry of a collimated beam was feasible and would provide useful information, but that some improvements should be made in the equipment. Accordingly, modifications were made with the following specific objectives:*? 1. ability to position and aim the apparatus at a selected source with great accuracy, 2. detector resolution good enough to identify individual nuclides among a multitude, 3. simplified data handling and analysis, 4. collimation adaptable to a wide range of source strengths, provisions for measuring spectra from selected spots during and im- mediately after power operation, 6. better calibration. °R. Blumberg, T. H. Mauney, and D. Scott, MSR Program Semiannu. Proger. Rep. Aug. 81, 18967, ORNL-4191, pp. 4044, 1°%. Blumberg, F. F. Dyer, and T. H. Mauney, MSR Program Semiannu. Progr. Rep. Aug. 31, 1968, ORNL-4344, pp. 36—40, 196. 'R, Blumberg, F. F. Dyer, and A. Houtzeel, MSR Program Semiannu. Progr. Rep. Aug. 31, 1969, ORNL-4449, p. 31. 14 By June 1969 these objectives had largely been met in the equipment de=- scribed below. 3.2 General Description As shown in Fig. 1.3, the fuel circulating system and drain tanks are situated in underground cells which, during operation, were covered by two layers of concrete beams with a thin stainless steel sheet between the layers. Gamma radiation levels in the reactor cell were 40,000 to 70,000 R/hr when the reactor was at full power, dropped to 3000 to 5000 R/hr upon a shutdown and drain, then slowly decreased.'?® Gamma radiation in the drain tank cell ran as high as 25,000 R/hr immediately after a drain.'? Thus, the situation dictated that any gamma spectrometry measurements would have to be made from a distance of 10 to 20 ft through apertures in a bio- logical shield. Even at the top of the shield, the intensity of the gamma-ray beam through an opening was quite high. For example, the beam above a 5-in.- diam hole in the portable maintenance shield, about 14 ft above the primary heat exchanger, was on the order of 500 R/hr one or two days after a shut- down and drain. Thus the radiation to the detector had to be reduced by collimation and sometimes by attenuation through shielding plates as well. Of course, the collimation of the beam was necessary also to restrict and locate the source of the gamma rays being analyzed. Figure 3.1 is a schematic, general view of the ultimate equipment, consisting of a collimator, a detector, and a laser alignment device. Figure 3.2 is a front view of the equipment. In these illustrations the equipment is mounted on the portable maintenance shield, but it could also be mounted over small holes drilled through the concrete shield blocks especially for this purpose. The detector was a Ge(Li) crystal connected through appropriate amplifiers to a 4096-channel analyzer. This combi- nation provided the high-resolution capability that was necessary. Dif- ferent collimator inserts could be used, depending on the intensity of the 12\, Houtzeel, MSR Program Semiannu. Progr. Rep. Aug. 31, 1968, ORNL-4344, pp. 22—23. 15 ORNL - OWG 68-139B0R2 LASER LASER ALIGNMENT BEAM GeHJ)DETECTOR_‘\\Ti COLLIMATOR INSERT COLLIMATOR BODY /,/ o L PORTABLE MAINTENANCE s s s i g SHIELD RAIL 48in.+ {2 BLOCKS REMOVED) NOTE: N ——— ‘ PORTABLE SHIELD ROLLS EAST AND WEST; CENTRAL PORTION ROTATES FOR NORTH~SCOUTH MOVEMENT VIEWING CONE —== HEAT EXCHANGER LINE 402 Fig. 3.1. Schematic arrangement of the remote gamma-ray spectrometer on top of the portable maintenance shield. Fig. 3.2. Front view of detector, portable maintenance shield. PHOTO 979004 LEAD GLASS WINDOW collimator assembly, and laser on [ ] & - 17 gamma radiation. When the equipment was used with the portable mainte- nance shield (PMS), the location from which the detected radiation was coming could be seen by shining the low-energy laser beam down through the collimator and looking through a lead-glass shielding window in the PMS. The precise location of the beam could be determined with a pair of sur- veyor's transits set up on the floor beside the reactor cell. 3.3 Detector and Amplifier The detector was a coaxial, lithium-drifted germanium detector mounted on a specially extended arm at a right angle to the cryostat. The manufacturer, ORTEC, measured a total resolution of 1.78 keV FWHM at 1333 keV at the time of delivery. The detector was 36.65 mm in diameter by 28.5 mm long with a total active volume of 26,25 ecm®. The peak-to-Compton ratio was 27/1. The measured efficiency (the ratio of the area under the 1333-keV photopeak recorded with this detector to the comparable area re- corded with a 3- by 3-in., Nal crystal, with 25 cm between the source to the detector) was 4.3%. Most of the experiments were done with an ORTEC-440A or ORTEC-450 amplifier, both of which were adequate and posed no problems. Although the collimated beam to the detector was very intense, it ac- tually interacted with only a very small portion of the total germanium crystal. Although some 1400 spectra were taken in a period of six months under far from ideal conditions, it did not appear that the system reso- lution deteriorated appreciably during the rugged handling. 3.4 Analyzer A Nuclear Data 2200 series, 4096-channel, analyzer was used. All spectra were recorded on magnetic tape. Initially several problems were encountered in operating the analyzer and tape unit, partly due to 'bugs" in this new system. Other problems appeared to be connected with the fact that the analyzer was initially located in a quite hot and very humid area and had to be shut down frequently for experimental reasons. Once the analyzer and amplifier were placed in an air-conditioned, humidity~controlled room and left on power continuously, most of the prob- lems disappeared. This move necessitated a 200-ft cable between the 18 detector preamplifier in the reactor high-bay area and the amplifier lo- cated near the analyzer; the long cable did not noticeably influence the system performance. System gain shifts, sometimes of several channels, seemed to a certain degree dependent on local power conditions (voltage and possible frequency variations). We accepted this fate and had no op- portunity to investigate this dependence. This high-resolution detector and 4096-channel analyzer combination performed satisfactorily. However, when the reactor was at power the cacophony of gamma rays from short-lived fission products in components such as the primary heat exchanger or reac- tor off-gas line generally proved too much even for this system; peaks de- generated into broad conglomerates from several gamma rays. In those cases, either a fair amount of shielding had to be used between the detector and the component observed to cut down the radiation level, or such a spectrum could not be analyzed. 3.5 Collimator Assembly Figure 3.3 shows the collimator assembly, which consisted of a col- limator body and a collimator insert. The collimator body, a lead-filled cylinder 32 1/2 in., high and 19 in. in diameter, had a central hole into which a collimator insert, also a lead-filled cylinder, could be placed, Three different, interchangeable collimator inserts having beam holes 1/16, 1/8, and 3/16 in. in diameter, respectively, were available to cope with the different radiation intensities. Ultimately only the inserts with 1/16~ and 1/8-in. beam holes were used to collect data, The inserts were 12 in. long and 3 7/8 in. in diameter. The straightness of the beam holes in the inserts proved to be a manufacturing problem since drilling of such small holes over a 12-in. length is almost impossible. Reasonably straight holes were obtained by placing thick-walled stainless steel precision tubes of 1/16, 1/8, and 3/16 in. inside diameter in the collimator inserts and then filling the inserts with lead. The detector and Dewar bottle were mounted on a platform attached to the collimator body. They could be moved on this platform with an adjusting screw to ensure that the detector was always at the same location over the collimator beam hole. The laser was fixed to the collimator insert with a 19 ORNL-DWG 69-8050 J LASER—— . i LASER ALIGNMENT BEAM — i Ge(Li) DETECTOR \H\ COLLIMATOR INSERT # 9 7 COLLIMATOR BODY é ? / -1 Fig. 3.3, Schematic of collimator assembly, detector, and laser. 20 small jig in such a way that the laser beam could be adjusted to shine through the middle of the collimator beam hole. For the laser to shine through the collimator hole, it was necessary to get the detector out of the way by moving the detector and Dewar backward on the platform with the adjusting screw, It would have been easy to guide the laser beam around the detector into the collimator beam hole with some sort of a mirror sys- tem, but it was thought that the anticipated rough handling of the detector, laser, and collimator assembly in moving it from place to place would not permit this extra complication. The collimator insert and laser (in fact, one subassembly) could easily be taken out and replaced or adjusted. 3.6 Setup of Experimental Equipment Several days were required after reactor shutdown to remove the upper shield blocks and the containment membrane, set up the PMS and the spec- trometry equipment, and start a survey. For this reason, it was decided to drill holes through the upper and lower shield plugs at three different locations in the shielding so that the spectrometer could be used while the reactor shield plugs were still in place. The three locations chosen were: one over the reactor system off-gas line, one over the primary heat ex- changer, and one over a drain tank. The holes were drilled during the June—August 1969 reactor shutdown period., The alignment of the holes in the upper and lower shield plugs was done with surveyor's transits. Of course, the containment membrane between the layers of shield blocks in both the reactor cell and drain tank cells remained intact. This setup, with the collimator assembly and detector placed on top of the upper shield blocks, was very useful for taking gamma-ray spectra during and immediately after reactor shutdown; it also gave us an opportunity to study the shorter- lived isotopes, at least at these three locationms. We used the PMS to survey the fission product activities along the 13 heat exchanger axis, the fuel lines, and the off-gas line, By placing the collimator assembly and detector over one of the holes in the rotating '?R. Blumberg and E. C, Hise, MSRE Design and Operations Report. Part X. Maintenance Equipment and Procedure, ORNL-TM=-910. 21 work plug of the shield, we could survey 35 in. of the heat exchanger with one setup. Figure 3.4 shows the general arrangement. By placing the PMS over any group of two adjacent lower shield plugs, we could survey large areas in the reactor cell, The collimator assembly and detector were placed on a large axial bearing so that the assembly could be turned freely on the PMS. This proved to be very useful for locational purposes. 3.7 Locational Equipment It was important to know precisely the locations on the different com- ponents from which gamma-ray spectra were taken. A low-power laser and two surveyor's transits were used for this purpose. The laser was mounted in the previously mentioned adjustable jig which could be attached to any of the collimator inserts. The laser beam coming through the center of the collimator hole and falling on the reactor component surveyed would then accurately indicate the center of the spot from which a gamma-ray spectrum was taken. The red laser '"dot" could be observed very easily through a lead-glass shielding window in the PMS (even on a 1/2-in. line some 15 ft away). Through previous calibration with a plumb line and an adjustable bubble level on the collimator body, we could ensure that the laser beam was perfectly vertical. With the surveyor's transits, placed in the re- actor high-bay area, it was then possible to locate the vertical laser beam and hence the spot examined on the reactor component., This rather simple system proved to be accurate within a fraction of 1 in. 3.8 Shielding As already mentioned, the radiation intensity from several components was rather high. With the reactor at power, the radiation levels from the holes in the concrete shield plugs over the reactor off-gas line and heat exchanger were more than 1000 R/hr. An important dose due to fast and thermal neutrons was also detected. (Except when spectrometry data were being taken, the holes were plugged and covered with lead bricks.) With the reactor at full power, it proved to be impossible to take meaningful spectra through the hole over the heat exchanger. So many PHOTO 97902A oy, RS B B CRY Fig. 3.4. General arrangement of detector-collimator assembly and portable maintenance shield placed over removed lower shield plugs. ) _) Zc 23 fission product nuclides appeared in the circulating fuel salt that among the great multitude of photopeaks present in these spectra, many coincided to form meaningless, large conglomerate peaks., These spectra could not be analyzed by our computer program for spectrum analysis. The off-gas con- tained fewer different nuclides than did the fuel salt, but even so, spectra taken with the reactor at full power through the hole over the off-gas line were on the borderline. A liberal use of shielding material, such as 2 in, of lithium-impregnated paraffin and 1 in. of copper, was necessary to render those off-gas-line spectra analyzable. (Lithium-impregnated paraffin was used since it absorbs neutrons without emitting capture gamma rays.) With the reactor at low power or right after shutdown from high power, we were able to analyze spectra from both the heat exchanger and the off- gas line. Still a heavy amount of shielding was necessary to keep the count rate from the detector within reasonable limits, 3000 to 10,000 counts/sec. A problem encountered was that many good shielding materials, such as lead, attenuate low-energy gamma rays (energies of a few hundred keV or lower) too strongly when enough shielding is used to obtain the desired degree of attenuation for the higher energy photons. Thus low-energy peaks could be entirely wiped out under such conditions. Because materials of low atomic number do not show such a strong shielding effect at low ener- gies, this problem was largely overcome by using shields made of aluminum and/or copper. Thus the count rate could usually be kept at a desired level while allowing measurement of the lower energy photons. Chapter 5, concerning calibration of the equipment, will show in more detail the prob- lems encountered. In general, we tried to operate the system at a dead time of 25% or less using a minimum of shielding material and the 1/16-in., collimator insert, 24 3.9 Calibration Setup The detector system efficiency was determined empirically, since the geometry, especially of the primary heat exchanger, was thought to be too complex to ensure adequate precision from theoretical calculations. The complete calibration procedure is given in Chapter 5. A subsequent check of this calibration using simplified calculational models is described in Appendix D. Figure 3.5 shows the physical setup for the calibration experiments. The detector, with a collimator, was installed in one of the underground cells (the decontamination cell) in the reactor high-bay area. A mockup of a complete section of the primary heat exchanger with a radiation source tube was located in the adjiacent remote-maintenance-practice cell. Radi- ation from the source tube positioned in the heat exchanger mockup passed through a specially drilled hole in the wall between the two cells and the collimator and then interacted with the detector. The wall between the cells was thick enough to eliminate any radiation at the detector except that coming through the collimator. The remote-maintenance cell was closed off with the regular roof plugs, except for the area directly above the heat exchanger mockup; the PMS was used here. Through holes in the PMS, we could then manipulate the source and the dummy heat exchanger tubes in the mockup with a simple tool. A 24-Ci ''°"Ag source was used for the calibration experiment., This source, activated by irradiation in the Oak Ridge Research Reactor, was 6.3 in. long and 1/2 in. in diameter. (It should be remembered that the real heat exchanger tubes also have a diameter of 1/2 in.) By placing this source tube in each of the heat exchanger tube po- sitions, with dummy tubes in all the other heat exchanger tube positions, we measured the influence of radiation from every one of these tubes on the detector. The addition of the readings from all the heat exchanger tube positions gave the detector response that would have been produced by a heat exchanger completely loaded with activated '*°"Ag tubes. The dis- tance between the heat exchanger mockup and the detector was made the same 25 ORNL-DWG 70-9563 HEAT EXCHANGER MOCKUP CRYOSTAT N\ LEAD SOURCE SHIELD Fig. 3.5. . . » ‘e . o ‘e LI - .o ' . "o ' . .. ' . [ 4 [ Con e L . SOt Ty S Plan view of calibration arrangement. \LASER DETECTOR COLLIMATOR 26 as the average distance between the detector placed on the PMS and the actual primary heat exchanger in the reactor cell. The heat exchanger calibration experiments were done with a dummy section of the heat exchanger outer shell (1/2 in. llastelloy N) placed be- tween the tube mockup and the detector. The alignment of the collimator hole between the source and the detector was again done with the laser rig, which was mounted on the now horizontally placed collimator insert. The red laser beam, passing through the collimator insert, was easy to locate and was aimed at the center of the heat exchanger mockup. The calibration required for the measurements on the reactor main off- gas line was less complicated because the portion of the line that was ob- served was simply a corrugated l-in. tube. It was estimated that the effect of this corrugated tube in comparison with the source tube would be approxi-~ mately the same on the detector. The ceramic part of a heat exchanger heater, as well as the shielding plates we used during the actual survey of the heat exchanger and off-gas line in the reactor cell, were calibrated with the same source to compare their shielding capabilities with those calculated from literature data. For this purpose, the ceramic heater was placed close to the heat ex- changer mockup with the source tube in an unshielded center position. The shield plates were placed as close as possible to the wall in the remote- maintenance practice cell in front of the hole. The shielding capabilities of the different plates as found by these tests agreed well with the values calculated from literature data. 27 4, ANALYSIS OF SPECTRA 4,1 Purpose and General Procedure The purpose of analyzing a gamma-ray spectrum is to identify radio- active nuclides by their emitted photons and, taking into account the ab- solute counting efficiency of the system, to determine the amounts of the various nuclides present at a given location in a particular reactor com- ponent. The counts in each one of the 4096 channels of the analyzer system are stored in the analyzer memory and, after the end of the counting time, written on magnetic tape if so requested. Hence, the basic data are counts assembled in channels. Each channel represents a particular energy interval whose size can be set more or less at will. We generally used an energy interval (so-called "gain') in the range of 0.3 to 1.0 keV per channel., 1In other words, we could take a gamma spectrum comprising a range of 1200 (v0.3 x 4096) to 4000 keV (1.0 x 4096). A gamma-ray peak (usually called a photopeak or full-energy peak) would identify itself in such a spectrum as an increase in accumulated counts in several adjacent channels. Experi- ments have shown that such a photopeak has essentially the form of a modi- fied gaussian curve. The intensity of a photopeak is determined from the height of, or more precisely from the area under, such a gaussian curve, The energy of the gamma ray is determined from the location of the gaussian curve in the spectrum, that is, the channel in which the top of the gaus- sian curve falls, The analysis of a recorded gamma-ray spectrum containing several photo- peaks proceeds through the following basic steps: 1. Locate the peaks in the spectrum, 2. Determine for each peak if it is just a statistical fluctuation or a real photopeak and determine a beginning and end of each peak. 3. Determine the background from adjacent channels for each peak. 4. Fit a gaussian curve to the net counts in those channels that form a peak. 28 5. Calculate the area under the peak as a measure of the gamma-ray intensity, subtract the background (above) from this area, and de- termine the location of the top of the gaussian curve to assign an energy to that peak. 6. Examine the spectrum peak by peak to determine the nuclides responsi- ble for the observed spectrum; a search is made through an energy 1li- brary to find those isotopes that might emit gamma rays within the range of each energy peak. 7. Select the nuclide most likely to be the one responsible for the peak in question. Each peak is subjected to several tests before it can really be associated with a certain isotope. These tests include half-life, presence of associated gamma rays and their relative in- tensities, and presence of precursor and daughter nuclides. 8. Once a peak is thus identified, calculate from the area under the peak how much of the particular isotope is present, A requirement for such a spectrum evaluation is that the efficiency of the detector system as a function of photon energy, that is, the ratio be- tween the counts detected and the gamma rays emitted, must be well known for the energy range considered. Also, the relationship between gamma energy and channel number must be determined to evaluate the energy of the peaks accurately. A typical spectrum is given in Fig. 4.1, 4,2 Problems Because most of our spectra were taken relatively soon after reactor shutdown, many short-lived isotopes were still present. This resulted in a large array of photopeaks, many of which overlapped each other. This means that several single photopeaks (singlets) would conglomerate (multi- plets); the analysis of these multiplets poses an entirely new dimension to the analysis of a spectrum. For example, the beginning and end of any one peak in a multiplet is very difficult to determine, as is the subtrac- tion of the background from this multiplet. Basically, a few gaussian curves have to be shifted back and forth so as to fit a multiplet. ORNL-DWG 72-7444 29 1050 I Y8901 e @ oateet -3 o _Ww 90t T e 8 . e T i B CO| s T o 2t ) w Dlopiay <. 3 ¢ Lo memeic®* 12E1 : P g anN—-fl-.H.- gl ‘Pfl&u..l“.lll.l..._lhu l.-”lll- T QL L e 3 4 B L k8 S - £ . g Tae ~ - | o ¥ ......“..a. . : ’ .*m n_b ” b s e s s b gl e : o 3 R i . 0 18 o h 1 e R S @ O . 5 NM—-;.‘“I-AII-!N.—.WD ] . T . . -.- -a- |II..—| . H . ..J.\ 7 |.|I|.|-|I-..'..-\- ..m m 2¢l ——— o o : - Iygmasmmm = ™ Y e 5 N i T Fep, N : S H__m_ - . Q” . I-llllfln.lfl-nm.\ ) HNm—. o -..q.l.nulll..n.l.n...-.t. " . 1= Lo — o Ii.l.l.....l.i«...,...f. ¥y . ) 0PE—.M-!H-FI]..II-II. ;“I . nf.qa.. o E 91 worky Yata o 18 & . gl m. * z, \.~.. W £ Ligy ey 4 e & 3] £ oL « oW mmllfl.rl..llluq-o.\.flnw u-..“.. :m nUu o e -, . ’ o= Tpgyerm L e x oF N ekl {a w w0 : ™ TN L e o a H__n_lf.l.u..mo o uPE_.m_illl.I...llrl\.H?. vlmmI'ln\.\ e’ 3 H‘ 3 &« Ww W WILE |yt T e it o it 3 e e R et taes, . e 4] Toet — Tos oL S gz M anmlll. = : — l.l.flw- . ..- - B = I gg ™=y ~ c ~F. = : g™ oW <% o g o W 3 Wee g LA 3 =z . ®Llact [ — T " SET g &0 ......... = s % xr HNmelnIEr.ll.l. IHfl.nl-..V .Hnn_.ifl.:-l = - o 5 1€} o fl.\.. . = QQ ) , o._.Een—r” , tfi.fl\ ...-w....s-.... . -w ; - Rflh. QFEemlfiI— —— .ll...".l...o ..m . . m t O g g s LTI s - TN @ :W 95, gyt & e Moo~ Il.m $- = o 1 A —— # 4 8 1 I Il.l...l.l.. ™ QL e L, _,.!n..v $2) .lrln....r CE |rmmantems T I iy, . 0 11 'y L€l 5! 1 R &t . L L o p— A —— 2E S e e *Weg ™ P 12¢) uemy " 1 H \laey )T, Ipp et ] @ - Zx Izg e - ] .M :mwo_.HNn_in.l.l.fl... - |m m e DL o e v s ” f nyg, o—..ll..i.-f- WigH ..H..._.-.. . | . ; “A‘ . 1o * o7 “l q.n- .. . L it ~ V- . \u :” 1 . M ~_... uJOv_lll..”.l - ...III!... - . " o fl a-..—-\- .3].-'- .y Im m o N 9 .f..,... o ) 2. ' 4% _ o g &) + . N e ® S I 13 0 ©, Weely, = © ¢ S l..u..". . N : 1 B P Gadaiais 1 L ad I o et s o e i 1 i Jalaiada i -l.-h..fll/t- 1 ” .c_r.".»"_ a_w. Lad 1 i - = & o - G © o e -t - w a* " ™~ - - o » ” o~ — - » - m ~ - L] w > ~ o~ Z 31vd LINNOD FAILYI3Y 30 The identification of isotopes by their emitted gamma rays is rela- tively easy if one has an extensive, accurate library of gamma-ray energies and branching ratios. The gamma rays of longer-lived isotopes are well known, but the shorter-lived ones are not and their descriptive data are scarce, As implied above, it is a difficult task to analyze one spectrum com- pletely, especially if it has to be done manually. If, as in our case, approximately 1400 spectra have to be analyzed, it becomes obvious that a computerized analysis is necessary. However, a computer program to handle our rather complicated spectra was not readily available, and it would have been too time consuming to have one written. 4.3 Table of Isotopes Considerable effort was expended in setting up a library of isotopic data. This library is probably fairly unique in that it contains both short- and long-lived fission product data and is taken from the latest references collected by the ORNL Nuclear Data Project. Gamma-ray energies and branching ratios are listed per nuclide as well as in order of in- creasing energy. Absolute gamma-ray abundances, as well as precursors and decay products, are also given. Use of this table proved to be a major improvement in the analysis of our spectra. It should be emphasized that the entire analysis of our data is based on this table of isotopes. The basic isotope table is given in Appendix A. As an example of possible errors, we might mention the problems en- countered with the analysis of the 129M7e data. First, this nuclide decays 129 129 partly to the ground state (" °“"Te) and then to I or directly from the 12971, In both cases, the abundance of emitted gamma rays isomeric state to is very small. This means any statistical error in the original photopeak analysis is amplified by a large factor for the activity calculation. Even more dis- turbing is the fact that the whole ***"Te-'2? Te decay scheme is not com- pletely agreed upon; a literature survey yielded several different decay schemes. Because the second largest photopeak (696.0 keV) observed in the 31 decay of '?®"Te coincides with photopeaks of other nuclides, we had to rely only on the results yielded by the 459.6-keV peak. Based on the latest available data, we estimated that the decay of 129Te and subsequently to *?®I would produce a 129MTe to the ground state gamma ray at 459.6 keV with an abundance of 4.97%. Should better data be- come available, the results presented later on should be corrected propor- tionately. The abundance of the 459.6~keV photopeak for the decay of *?°Te 129 itself to I is estimated to be 7.7%. 4.4 Computer Programs As already mentioned, computerized data handling was imperative for the proper analysis of this experiment. The principal problem was where to find a computer program that would analyze our spectra with their multi- tudes of singlet and multiplet peaks. As an indication of the complexity, it should be emphasized that we commonly had 150 to 230 peaks per spectrum. The ORNL Mathematics Division had developed an effective gamma-ray spectrum analysis program, GAMSPEC-3, which was operational but did not yet have provisions for analyzing multiple peaks. The program was also limited to the analysis of 75 single peaks per spectrum and would not identify nu- clides., These limitations eliminated this program for the analysis of most of our spectra, especially those still containing many photopeaks from short-lived isotopes. Gunnink et al,,'* among others, had developed a versatile gamma-ray spectrum analysis program, GAMANAL, which seemed to suit our purposes. This program had been written in a Lawrence Radiation Laboratory (LRL) version of FORTRAN for a CDC 6600 computer and had been converted to FORTRAN-IV and made compatible with the IBM 360 series of com- puters by N. D. Dudey et al., of Argonne National Laboratory. Through a concerted effort of the ORNL Mathematics Division, the program was made compatible with the ORNL computers. As an indication of the complexity of '“R. Gunnink et al., Identification and Determination of Gamma Emitters by Computer Analysis of Ge(Li) Spectra, LRL-UCID-15140, 32 the program, it might be mentioned that it took LRL close to five man-years to write this program, ANL close to two years to make it compatible with the IBM 360 computers, and ORNL some four months to get it running here. This program will locate peaks (singlets, doublets, and triplets) and identify and calculate the amounts of nuclides responsible for the peaks. It can handle 400 singlet peaks, 100 doublets, and 100 triplets. Although the basic program was originated by Gunnink et al., we made some modifi- cations to it and supplied our own library of isotope data. For example, the original program read in the efficiency of the whole detector system as a function of energy as coefficients of a polynomial equation. This was modified to have the program interpolate efficiency values from a set of data that represented the efficiency curve itself. Normally we supplied approximately 60 points on this efficiency curve. It was also changed to permit a larger choice in output of different tables in order to reduce the paper output, Although this program worked satisfactorily for most of our spectra, some problems remained, as follows. 1. In some specific energy ranges, many peaks appeared in multiplets, Since the program would only analyze up to triplets, it was obvious that a conglomerate of five or six peaks could not be properly analyzed. 2, The storage possibility of 100 doublets and 100 triplets some- times was inadequate for spectra that contained many short-lived isotopes. Such a spectrum would then be rejected. 3. It was only exceptional that all the nuclides of a particular de- cay chain were deposited. When the program then went through the different identification tests, it would sometimes fail to find the precursor or daughter products and hence reject an identification. The only way out was then to uncouple the parent-daughter relation in the energy library. For example, in chain 95, ®°Zr remained with the fuel salt while ®3Nb tended to plate out. The program would not identify ?°Nb as long as ®°Zr was coupled to it as its precursor. Of course, this had nothing to do with the per- formance of the program itself. 33 5. CALIBRATION 5.1 General The equipment was calibrated to determine the absolute efficiency of the whole gamma-ray detection system, taking into account the collimator assembly and the physical arrangement of the equipment in the reactor cell. The counting efficiency as used in our empirical method is merely a proportionality constant relating photopeak count rate and source inten- sity, viz., CR = EF +« S , (1) where (R is the count rate, FF is the efficiency, and S is the source in- tensity expressed as photons per unit of source dimension and per unit time. It should be notedvthat even though CR is determined by photons from the total radiocactive object subtended by the collimator, the source strength can be expressed as any convenient fraction or multiple of the total object. For example, in the case of the heat exchanger (see Sect. 5.5), count rates that corresponded to the collimator subtending a conical section of the heat exchanger full of radioactive tubes were obtained. The source strength, however, was taken as those photons emitted per unit time per square centimeter of a single tube. Thus the counting efficiency is expressed as e CR(counts/min) 9 @) Y/cm 5( Y ) cm? — min FF( where y denotes an emitted photon and ¢ a count registered in the photopeak. When a count rate measured over the heat exchanger was divided by this ef- ficiency, the result was the photon emission rate per square centimeter of tubing in the heat exchanger. 1In the case of the reactor main off-gas line, the source strength was expressed in units of photons emitted per inch of tubing per minute (y/in.-min), causing the efficiency to have the units 34 c v/in. energy of the incoming gamma rays. Basically, there are two ways to de- The efficiency of any system of this kind is dependent on the termine the relationship between efficiency and energy. One could estab- lish this efficiency curve by measuring first the detector efficiency in a simple geometry with several different known sources and then discount the effects of collimation, distance, shielding, and geometry for an actual component to be surveyed. The other way would be to establish empirically the efficiency of the system in the actual physical setup. We chose the second path because it appeared to be somewhat simpler and more reliable. However, we also performed our analysis using the more basic approach to check the validity of the calibration (see Appendix D). All gamma spectra taken from July to September 1969 were obtained with the ORTEC-440A amplifier. During our calibration experiments, we changed to the ORTEC-450 amplifier. Because it was not known at that particular time what the influence would be on the resolution of the system, it was decided to do a fair amount of the calibration work in duplicate using both ampli- fiers. 5.2 Source In order to simulate the actual conditions as much as possible, we chose a 1/2-in.-0D silver source tube that was equivalent in diameter to the primary heat exchanger tubes. Silver was chosen because '*°"Ag has several gamma rays between 446.8 and 1562.2 keV and thus yields data over a sizable part of the energy range for the efficiency curve. Furthermore, it was rather inexpensive to prepare such a source by using the hydraulic-tube irradiation facilities of the ORR. Because of the large effect of the silver tubing on the reactivity of the ORR, only 2.3 in. of tubing could be irradiated in a single hydraulic tube; so three pieces of silver tubing of this length were irradiated, each in separate hydraulic tubes. Flux monitors (Co-Al foils) were first irradiated to determine the flux gradients and total flux in the hydraulic tubes and provide a basis for the silver irradiations. The assembly of our source tube was done in the hot cells of the ORNL Isotopes Division and consisted 35 in putting the three silver tubes together in one thin-walled stainless- steel container and seal welding it (Fig. 5.1). Appendix B reports in more detail the calculations done concerning the source strength as well as the activity gradient along the length of the source. 5.3 Slit Experiment — Source Strength Because of the neutron flux distribution in the ORR, it was inevitable that there was some activity gradient along the source tube., It was neces- sary to know this gradient to properly calibrate the detector system. Originally we planned to deduce the level of '*°"Ag radiocactivity at any point on the tube from the measured activity values and the activation gradients obtained from the flux monitors (Table B.2 and B.4 of App. B). However, this procedure has many uncertainties. A much better procedure, and one that was followed, was to scan the source with a collimator ar- rangement using the Ge(Li) detector and 4096-channel analyzer. A slit shield, set up in the remote-maintenance practice cell at the MSRE, con- sisted of stacked lead bricks to a height of about 2 ft. An air-gap slit 1/8 by 1 in. wide was formed in the brick stack by using 1/8-in. lead spacers; a glass tube large enough to accept the source was attached verti- cally to the brick stack over the air gap. The Ge(Li) detector was placed in the adjoining decontamination cell in front of the hole drilled through the concrete wall. When the source was placed in the glass tube, gamma rays from a 1/8-in. section of the silver tube could pass through the slit, the hole between the cells, and the collimator to strike the detector. Scanning was begun by lowering the source below the slit and then gradually raising it until the detector signaled that the top of the source was in front of the slit. The scan was made by moving the source upward and obtaining gamma spectra at 1/8-in. increments until the entire source had been moved past the slit, A profile of the source intensity was obtained by resolving the gamma spectra and plotting net counts of photopeak areas versus distance along the source (see Fig. 5.2). From these results we could then calculate the absolute activity at any point along the entire length of the source tube (App. B). 36 ORNL-DOWG 70-9564R T . §.3 jp, - —w{ =Y in g 72 J 7 L L L ol a Z £ L 7/11/{}//4////// [[J[I/J]_er[ VN . N . N N NN N S S N N S N8N NS N NN 0.5in. ) {[_,217/—7_LL11[LL ////LL/ y / PLUG SOURCE TUBE PLUG BAIL END CAP Fig. 5.1. Source capsule assemblyv. _ . ORNL-DWG 70- 9565 35,000 : I | i MEASURED ACTIVITY 4,196 Ci/in. , - | | LA e b s, 30,000 = | 1 MEASURED ACTIVITY 3.861 C|/|n ~T 25, 0Q0 : ! . 4!_ 20,000 l ! ! e b e £ ‘ . . £ .‘ ' | 2 ‘ o | ’ | 1. g 15000 r——— +—4 t— +—-—7;%_:_?4 , ‘ | Hom i | ‘ A 657.7 k A 10, 000 ey b . &6 ev g o 1 o 0 884.7 kev ''0™aq J | | | | { { . ; | i | ! © 937.5 kev '¥™Maq | 1 | i 5000 - f - V( | | — ‘ , ( f | 1 | j ! - & : wi 0 — ) : ‘ | . i l . ; | | 2 TOTAL SOURCE LENGTH: 6.3 in. Fig. 5.2. Net counts of 3 major '!o™ Ag photopeaks versus distance along the source tube. LE 38 Because the activity profile of the silver source tube was not com- pletely uniform and the collimator subtended a significant portion of the tube, an average value of the source strength as viewed by the system was needed. The average source strength, S can be expressed as L f S(x) RE(x) dx 5§ = 2 , (3) L [ R(x) dr 0 where I is the length of the tube, S(x) is the source strength at point x, and R(x) is a weighting function equivalent to the radial counting effici- ency (sensitivity response to a point source as a function of distance from the source-to-collimator axis). Since the source strength S(x) and the count rate CR(x) found in the slit experiment are linearly related, we can write L k [ CR(x) R(x) dx 5 - — , ) L f R(x) dx 0 where k¥ is a proportionality constant relating S(x) and CR(x). Note that k is the reciprocal of the counting efficiency for the source in the slit experiment configuration and has the units of (y/in.)/c. We can now write L _ '([). CR(x) R(x) dx — S _ Ck = z’ - . ’ (5) [ R@) dx 0 where CR is the average count rate for each of the gamma rays for the total source tube. 39 In determining the weighting function R(x), we assumed that the col- limator-detector assembly had rotational symmetry; that is, only radial displacement from the collimator axis was considered. (This assumption required that the collimator hole be perfectly straight, which seemed rea- sonable, at least for the 1/8-in. collimator.) From the calibration work in the heat exchanger mockup (see Sect. 5.5), we were able to measure the contribution to count rate at the detector from sources to the right and left of the collimator axis. Figure 5.3 is a nor- malized plot of the detector sensitivity to the silver source tube at var- ious distances from the axis. This figure was made up as follows: For three different '*°7Ag photon energies, the net count rates were added for all the source tube positions in each column of tubes in the heat exchanger mockup (parallel to the collimator axis, Fig. 3.5). The reason for summing the readings of an entire column was to obtain a more uniform source dis- tribution as well as better counting statistics. This count-rate distri- bution across the mockup was then normalized to a central value of unity for each of the three photopeak energies and corrected for a slight mis- alignment of the collimator. Figure 5.3 is the average of the three nor- malized curves; their deviation from this average was encouraging small, The figure also represents the required weighting factor R(x). By multi- plying the count rates along the source tube, as measured by the slit ex-— periment (Fig. 5.2), in the appropriate way with the normalized count-rate distribution as seen by the detector (Fig. 5.3), the result is the effective activity of the entire source tube as seen by the actual detecting system (Fig. 5.4). The average weighted count rate of the source tube for the three energies considered is the area under each of the curves on Fig. 5.4 divided by the area under the normalized curve (Fig. 5.3). For the three energies we find the following average weighted count rates: Ener (keV Count rate (counts/min) 657.7 27,500 884.7 24,300 937.5 13,600 Since we now know the average count rates for the unshielded total tube for three of the gamma rays of the source, we need only calculate the RELATIVE COUNT RATE 0.9 0.8 0.6 0.5 0.3 ORNL-DWG 70- 9566 |“ TOTAL SOURCE LENGTH 6.3:in. : .e! Fig. 5.3. Normalized detector sensitivity of gamma rays coming from a source 15 ft away through a 1/8-in. collimator to the detector. 0% COUNT RATE {counts/min) ORNL-DWG 70-9567 7 30,000 ! ‘ w ! ‘ ‘ | | : ; | | l 10m A . 25,000 - + L E*_L _‘_‘ ] \ ! o J 20,000 | - : | 1 | | - ! | ’ o 15,000 - : . + ' | { | | | | ! - | [ | o 10,000 — - - | | . | | | | ! S R - . l ‘ | 5000 b —— S 937.5 kev”omug — . h l | 0 ' ] ‘ ‘l r ‘ L* -~ ———————— TOTAL SOURCE LENGTH: 6.3 in. e — - Fig. 5.4. Measured count rate from slit experiment multiplied by normalized count rate distribution as seen by detector. 42 value of k to determine the average source strength. The value of kX can be determined from the expression __S(x) K& = Tag where S(x) and CR(x) denote the source strength in curies per inch and counts per minute, respectively, at point x on the tube as determined by the slit experiment. The subscript 7 denotes the 7th gamma and accounts for the variation of % (a slit-experiment counting efficiency) with energy and the fact that the three gamma rays measured each have different branch- ing ratios (photons emitted per 100 disintegrations). Since we know the source strength at two positions (see Fig. 5.2 and App. B), we can use the results of the slit experiment to obtain the value of Xk at two positions for each of the three gamma rays as shown in Table 5.1. Table 5.1. Count rates measured at two positions along the silver source for three different photopeaks Count rate from Count rate from Photon slit experiment slit experiment _ energy for 3.861 Ci/in. k(x 10%) for 4,196 Ci/in. k(x 10%) K (keV) (counts/min) _(Cifin. (counts/min) _(Cifin. counts/min) counts/min) 657.7 25,900 1.491 28,300 1.481 1.486 x 10™° 884.7 23,100 1.675 25,100 1.671 1.673 x 10™° 937.5 12,700 3.044 14,700 2.861 2.953 x 107° We have listed the count rates and values computed for k for each of the gamma rays for both "known'" positions of the source tube. In addition, we have listed for each of the gamma rays the average value of k for the two positions of the tube. Three values of the average source strength were then calculated by multiplying these averaged k values times the cor- responding averaged count rates for the three gamma rays given above. The results of these calculations are: 43 Ener keV Source strength (Ci/in.) 657.7 4,09 884.7 4,07 937.5 4,00 From these results an overall average source strength was computed to be 4,05 Ci per inch of tube on June 1, 1969, This source evaluation method leaves something to be desired and is to a certain degree misleading because we used an essentially plane source (source placed in all heat exchanger mockup positions) to evaluate the weighting factor of a line source., However, since the normalized weighting curve (Fig. 5.3) 1s quite steep and the source activity gradient in the central portion of the source rather flat, the method should be acceptable. Taking into account the errors of the radiochemical source-strength evalu- ation and the weighting-factor evaluation, we estimate an error of 57 in the source-strength computation. 5.4 Single Source Experiments A single source experiment was performed to determine the efficiency of the detector system for gamma rays coming from the main reactor off-gas line. We .assumed that the detector system would have the same counting efficiency for the **®"Ag source tube as for the corrugated 1-in.-ID main off-gas line. The source tube was set up in an unshielded central position of the heat exchanger mockup, and the collimator was aligned with the laser. Minor collimator adjustments were made to cope with any misalignment by searching for the maximum count rate for this setup. The distance between the source tube and the detector was approximately equal to the distance between the actual detector-collimator assembly on the PMS and the main reactor off-gas line (line 522) in the reactor cell. Four spectra were taken for this setup with the 440A amplifier and six with the 450. The difference in response was small and judged to be due to statistical error. A 1/8-in. collimator insert was used. By taking into account the data of this experiment and the absolute abundance of the different gamma peaks (15 photopeaks of **°"Ag), we were able to calculate the detector-system efficiency at all these different 4 photon energies. This way we established an absolute efficiency curve from the lowest ''°"Ag peak energy at 446.8 keV to the highest at 1562.2 keV. The problem was to extend this curve beyond these limits, because many fis- sion products have important photopeaks outside this energy range. It was decided to use the actual fission product spectra taken from the off-gas line for the extension of the efficiency curve. For example, with *°Mo, 131 132 I I , , and *“°La, it would be possible to establish a relative effici- ency curve all the way from 140 up to 2522 keV. By linking this wide- energy—~range relative efficiency curve to the already established absolute curve from ''°MAg, an absolute efficiency curve was obtained. It was very encouraging to see that the relative curve from the fission products and the absolute curve were almost identical in shape in the 450- to 1560-keV range. Figure 5.5 is the efficiency curve adopted for the main reactor off-gas line (line 522). As a check on the above approach, we used also a 22¢ Ra source, which emits photons over a wide energy range. The relative efficiency curve for that source was essentially identical in shape with the established abso- lute curve. The discrepancies, especially in the lower energy range, were probably due to a different shielding of the actual photon emitter (for the l1-in.-ID off-gas line, this would be the wall thickness of the pipe). 5.5 Heat Exchanger Calibration The heat exchanger calibration was done with a full-size mockup of a complete section of the heat exchanger. The 8 1/2-in.-~long dummy tube sec- tions had the same outside diameter and wall thickness as the actual heat exchanger tubes. In the mockup the tubes were held vertical and kept at the specified triangular pitch of 0.775 in. by two horizontal grid plates approximately 6 in. apart., With a simple tool it was easy to handle these tubes from above through a work hole in the portable maintenance shield. The mockup contained 20 rows of tubes (transverse to the detector-collimator axis). Because of the triangular pitch, each row contained either nine or eight tubes. A curved 1/2-in. stainless steel plate was placed in the proper geometry in front of the mockup to simulate the heat exchanger shell, DETECTOR SYSTEM EFFICIENCY 300 500 700 Fig. 5.5. DETECTOR SYSTEM EFFICIENCY = ! ORNL-DWG 70-9568 COUNTS REGISTERED PHOTONS EMITTED PER LINEAR INCH OF TUBE 900 {100 1300 1500 1700 1900 2100 2300 2500 2700 PHOTON ENERGY ( kev) Absclute efficiency of detector system for photons emitted from the 522 line. 2900 ah 46 Spectra were taken with the source tube placed in each of the tube positions and dummy tubes in all the other positions. This procedure was done twice, once with the 440A amplifier and once with the 450; the 1/8-in. collimator insert was used. Some spectra were taken with the 1/16-in. col- limator, but a full calibration was not performed. By adding the count rates for each of the different *'°"Ag photopeaks from the spectra of all the source positions in the mockup, a count rate was obtained that represented the primary heat exchanger with an activity of 4,06 Ci of **°MAg per inch on each tube, but with the internal shielding of all the other heat exchanger tubes. By taking into account the data of the experiment and the absolute abundances of the various 1197MAg gamma rays, we could establish an absolute efficiency curve applicable to the heat ex- changer for the energy range from 446.8 to 1562.2 keV. Again, there was very little difference between the data obtained with the two different amplifiers; we used the average of the two efficiency curves as our standard. Actual fission product data from the primary heat exchanger were used to extend the energy range of the efficiency curve; we used the photopeaks of *°Mo, '*'I, '32I, and, to a lesser degree, '“°La (because very little '“°La was present in the empty heat exchanger). The shielding effect of a heater element was measured and also calculated separately. The final standard efficiency curve for the heat exchanger, including the heat ex- changer shell and the heater box effects, is given in Fig. 5.6. It should be noted that this curve, contrary to that for the off-gas line, bends down at lower energies. The efficiency decrease at low energy was caused by the internal shielding of the heat exchanger; hence the strong increase in the photon attenuation coefficient with decreasing photon energy. This down- ward trend with decreasing energy proved to be a problem in the initial use of the computer program (see Chap. 4.4). The above calibration procedure does not take into account the effect of fission products deposited on the inside of the heat exchanger shell. However, this amount, if comparable with the amounts deposited on the tubes, would have only a minor influence on the total reading. Another limitation was that the calibration would be reliable only if the deposition of fission DETECTOR SYSTEM EFFICIENCY 10 {0 -7 300 500 DETECTOR SYSTEM EFFICIENCY = 700 Fig. 5.6€. ORNL-DWG 70-9569 COUNTS REGISTERED PHOTONS EMITTED PER cm? OF HEAT EXCHANGER TUBE b 900 100 {300 {500 4700 {1900 2100 2300 2500 2700 2900 PHOTON ENERGY (kev) Absclute efficiency of detector system for phctons emitted from the primary heat exchanger. LY 48 products were approximately uniform on all the tubes. Although this is not necessarily true, we had no data to justify any other approximation. As a check on our empirical efficiency calibrations, we have computed from first principles counting efficiencies for both a line source corre- sponding to the silver tube and the reactor off-gas line and a volume source (truncated cone) corresponding to the cone subtended by the colli- mator in the heat exchanger. These calculations are described in Appendix D. We believe that the good agreement found between experimental and com- puted counting efficiencies serves to confirm our measured values and in- creases our confidence in the validity of our measurements. 5.6 Calibration of Shielding Materials Throughout the experiments we tried to keep the count rate of the de- tection system more or less constant in the range 3000 to 10,000 counts/sec. Since the source strength varied with time and position, we varied the detection-system efficiency accordingly by interposing different amounts of shielding material between the source and the detector. In most cases, disks, 1/2 in. thick and 7 in. in diameter, were placed over the work hole in the PMS just under the collimator assembly. These disks could be easily removed to permit use of the laser for location purposes. We used disks of aluminum, copper, and lithium-impregnated paraffin. With some extremely high activity levels, we also used some 1l/4-in.~thick plates of lead be- tween the collimator insert and the detector. Aside from the very incon- venient tendency for lead to attenuate especially well the lower energy photons, the proximity of this shielding material to the detector also caused the detection of lead x rays. In evaluating the influence of the different shielding materials, we relied not only on the published attenu- ation coefficients but we also calibrated the disks. We used both the un- 226pa source for this purpose. Our shielded silver source as well as the experimental values agreed well with the literature data for Pb, Cu, Al, Cd, and steel; we had to rely on experimental values only for the lithium- impregnated paraffin and the ceramic heat exchanger heater element. So- called "shielding curves" were made for each of these different shielding 49 materials; these curves were plots of the shielding factors (reciprocals of the more usual attenuation factors) as functions of energy from 140 to 2500 keV, Absolute efficiency curves as functions of energy were synthesized for all the different combinations of shielding material used in the ex- periments. These were obtained by dividing, point by point, the unshielded efficiency values by shielding factors appropriate for that shielding com- bination. When two or more shielding disks were involved, composite shielding factors were used. That is, EF (E) BE(F) = —H—— nia, 1" where EFS(E) = energy-dependent efficiency with shielding present, EFM(E) = energy—-dependent efficiency without shielding, Ai(E) = shielding factor for the 2th component of the shield, Ni = number of pieces of component % used. Altogether we used 12 different combinations of shielding for the main re- actor off-gas line and 8 for the heat exchanger (see Tables 5.2 and 5.3). The values of the efficiency curves generated for these shielding configu- rations were used as input data to the computer analysis program to evalu- ate the absolute amounts of nuclides present. The different efficiency curves are given in Appendix C, With the reactor system shut down and drained, we used primarily the simpler cases of shielding with only aluminum or copper. With the reactor at different power levels, we had to rely on much more shielding because of the high gamma and neutron radiation levels. Gamma spectra taken at the same place but at different times after reactor shutdown and recorded with different shielding configurations yielded generally the same results for the amounts of nuclides deposited at shutdown time; this gave us a certain degree of confidence in the absolute efficiency curves used. 50 Table 5.2, Shielding configurations employed during surveys of reactor off-gas line Collimator Shielding insert case diameter (in.) Shielding material 1 1/16 1/8 in. steel 1/8 1/8 in. steel, 1 in. Cu 3 1/16 1/8 in, steel, 1 in., Cu, 2 in. paraffin (Li), 1/8 in. Cd 4 1/8 1 in., Al 5 1/8 None 6 1/16 None 7 1/8 1/8 in. Cd 8 1/8 1/8 in. Cd, 1/8 in. steel 9 1/16 1/8 in., Cd, 1/8 in. steel 10 1/16 1/8 in. Cd, 1/8 in. steel, 2 in. paraffin (Li), 1/2 in. Pb 11 1/16 1/8 in. Cd, 1/8 in. steel, 2 in. paraffin (Li), 1/4 in. Pb 12 1/16 1/8 in. Cd, 1/8 in. steel, 2 in. paraffin (Li) 51 Table 5.3. Shielding configurations employed during survey of primary heat exchanger Collimator Shielding case insert Shielding material diameter (in.) 1 1/8 None 2 1/8 1 in, Al 3 1/8 1/2 in. Al, 1/2 in. Cu 4 1/8 1l in. Al, 1/2 in. Cu 5 1/16 None 6 1/16 1/8 in. steel 7 1/8 1/8 in. Cd 8 1/8 1/8 in. Cd, 1/8 in. steel Extensive experiments were done to evaluate the effect of the 1/16-in. collimator insert in relation to the 1/8-in, insert., During the calibra- tion experiments, we found values appreciably higher than 4 for the decrease in count rate when the 1/16-in. insert was used; also, there was consid- erable scatter between individual measurements. It might have been that this spread was largely due to minor misalignments between the collimator axis and the center of the silver source tube. Therefore, another experi- ment was done, this time by placing a lightly activated foil alternately against the far ends of the 1/8- and 1/16-in. collimator inserts and the detector placed at the front end. We found a shielding factor of 9.0 for the 1/16-in. collimator insert in relation to the 1/8-in., insert. This test was repeated several times, and the experimental values were within *10%. Apparently the 1/16-in. collimator was not all that straight! 52 5.7 Calibration for Fission Gases in the MSRE Because our calibration was done in terms of radioactive sources de- posited on the coolant salt tubes in the heat exchanger, all results ob- tained by the computer program, including those for gases, were expressed in this manner. Although this "equivalent surface deposition' for gases is a valid* mode of expressing their concentrations, it may for certain purposes be more useful to express their concentrations in terms of free cross-sectional volume of the heat exchanger. We have estimated the av- erage ratio of free volume to surface area to be 0.55 cm®/cm®. Thus the results for gases can be converted to disintegrations per cubic centimeter per minute by dividing their values in disintegrations per square centi- meter per minute by 0,55, % The validity of this mode of expressing gas concentration is easily seen from the arguments in App. D. 53 6. MEASUREMENTS Gamma-ray spectra were taken during the period of July 1969 through December 1969. Altogether some 1400 spectra were recorded from several different locations (Table 6.1). The purposes of the various sets of measurements were as follows. Group A — The reactor system had been drained and shut down since June 1, 1969. The purpose of these spectra was to test the equipment and, if possible, to study the deposition of longer-lived fission products, Group B — During this period the reactor was usually at some power level ranging from a few kilowatts to full power. By changing the reactor conditions, such as fuel pump speed, helium purge rate in the fuel pump bowl, and changing to argon as a purge gas, the concentration of the noble fission gases was studied; of special interest was, of course, '*®°Xe,. These spectra were all taken over a hole in the shield plug with the con- tainment membrane in place and were considered in a separate report. Group C — Calibration spectra were mainly taken with the *'°"Ag source tube; some were taken with a *?®°Ra sourxce. Details of the calibration work are described in Chapter 5. Group D — A few spectra were taken through the shield plug over the main reactor off-gas line during a beryllium addition to find out if the noble fission gas concentration changed during and after an addition. These results were also considered in Ref. 15. Group E — Several spectra were recorded of the main reactor off-gas line several hundred feet downstream of the fuel pump bowl. These spectra were taken at a location where sample bombs of fission gases can be iso- lated. Physically this location is in the vent house. The connection of the off-gas line into the main charcoal beds is a few feet downstream from this point. The purpose was to study the relative noble fission gas con- centrations at this point. 157, R. Engel and R. C. Steffy, Xenon Behavior in the Molten Salt Reactor Experiment, ORNL-TM-3464 (October 1971). 54 Table 6.1, Gamma-ray spectra recorded July to December 1969 523 line roughing filters Number Group Place Time of spectra MSRE heat exchanger July 114 Main reactor off-gas line (522 jumper line) 20 Main fuel line (line 102) 5 Through shield plug hole over off-gas line Aug.—Sept. 235 Through shield plug hole over heat 15 exchanger Calibrations (Chap. 5) Oct. 425 Through shield plug hole over off~gas line Oct. 9 (during Be addition) Main off-gas line in vent house Oct. 6 Through shield plug hole over heat Nov. 18 exchanger Through shield plug hole over off-~gas line 24 Through shield plug hole over drain tank 15 MSRE heat exchanger Nov. 241 Main reactor off-gas line (522 line) 91 Fuel pump bowl, fuel lines (lines 101 36 and 102) Through shield plug over heat exchanger; Dec. 43 final shutdown Gas samples, reactor cell Dec. 9 MSRE coolant salt radiator Dec. 5 Main coolant line Dec. 14 Samples of fuel salt and graphite Dec. 45 Miscellaneous: lube oil system, July—Dec. 10 55 Group F — From previous experience it was known that it takes at least a few days from the moment of reactor shutdown and drain until the portable maintenance shield is set up and operable over the reactor cell. In order to get some data during this time, spectra were taken over the three holes (off-gas line, heat exchanger, and drain tank) in the shield plugs. Reactor shutdown and drain on November 2, 1969, at 1441 hr was initiated by thawing the drain line freeze valve with the reactor still at full power. It was judged that this type of drain would leave the maximum fission product de- position in the system because the time span elapsed between the moment of fuel drainage of the entire fuel system and full-power operation would be shortest. Group G — These are the main spectra taken with the detection equipment mounted on top of the portable maintenance shield after reactor shutdown. Group H — Several spectra were taken from the heat exchanger over the hole in the shield plug. Spectra were recorded during full-power operation before final shutdown, at shutdown, during draining of the system, and thereafter. Group I — Spectra were recorded from samples of the reactor cell air after final reactor shutdown. These spectra were used in the analysis of the piping leak that occurred after the shutdown.®® Group J — Spectra were recorded of the coolant-salt radiator after the system was drained; the detector was placed in front of the radiator tubes. The purpose was to find any activated corrosion products that might have settled in the radiator., Group K = Spectra were taken by C. H. Gabbard at two places along the main coolant salt line to study the coolant flow rate by using the decay of 2°F and '°N in the coolant salt; *°F and *°®N are formed in the heat ex- changer by delayed neutrons from the fuel salt. Group L — Samples of fuel salt and graphite soaked in fuel salt were taken from the fuel pump bowl and placed in the sampler enricher. Spectra were taken by C. H. Gabbard with a special collimator and detector setup. '®R. H. Guymon and P. N. Haubenreich, MSR Program Semiannu. Progr, Rep. Feb. 28, 1970, ORNL-4548, p. 14, 56 Group M — Several miscellaneous spectra were recorded, such as from the lube-oil system, the roughing filters from the containment ventilation air exhaust, and from the off-gas line of the pump bowl overfiow tank (line 523). The results of the analysis of the different spectra will be reported by group in Chapter 7. 57 7. RESULTS Presented in this chapter are all results which pertain to the distri- bution of fission products in the MSRE. Brief mention is made of other spectra, recorded for different purposes, but their analyses will be dis- cussed elsewhere. Results are reported in the order given in Chapter 6. Because of the large number of spectra, results have been presented, where possible, in graphic form. 1In general, all results have been included in the figures, irrespective of possible experimental errors. Results that are somewhat uncertain for a particular reason are specifically indicated. Only in very few cases, because of obvious error, were data omitted. 7.1 Group A Spectra These spectra were recorded during the July shutdown period and con- cern the primary heat exchanger, the main off-gas line, and a fuel salt line. The shakedown of the equipment proved to be very useful, in that problems with the analyzer and detector system were resolved; the alignment and locational equipment, including the laser, proved to be sturdy and reliable, Most spectra were taken six or more weeks after reactor shutdown (June 1, 1969); this implies that only a few longer-lived isotopes might be expected to be present. The spectra were relatively simple and contained no multiplets. The system had not been flushed with flush salt, For nuclides emitting photons at different energies, we used the av- erage of the computed activities yielded by the most prominent photopeaks emitted by that nuclide. The counting time was mostly 200 sec per spectrum. Although '*7Cs and *“°Ba-La were identified a few times in the heat exchanger, their ac- tivity level was so small that one might tentatively conclude that they were deposited as a result of the decay of xenon present in the system after shutdown and drain. No ?°Zr could be detected. It should be borne 58 in mind that, except where specifically noted, all results in this section have been extrapolated back to reactor shutdown time by simple exponential extrapolation. The reason for this extrapolation was to compare results yielded by spectra taken at different times. This extrapolation might give, however, an activity value at shutdown time that is too high because of the decay of a precursor nuclide. For example, the extrapolated value of *®'I might be too high by a few percent because of the decay of '*'Te., This will be further discussed in Chapter 8. 7.1.1 Heat exchanger About 65 spectra were taken along the longitudinal axis of the heat exchanger, Another 49 were recorded by moving the detector transversely across the heat exchanger; the transverse scans were taken at four differ- ent places. Most spectra could be successfully analyzed. The quantitative interpretation of the transverse-scanning results proved to be rather dif- ficult, due to the varying shielding condition because of the changing num- ber of heat exchanger tubes seen by the detector as well as the changing photon attenuation through the heat exchanger shell. Qualitatively, the fission product deposition was symmetrical on both sides of the longi- tudinal axis of the heat exchanger. The spectra recorded above the heater connector boxes (HTR plug) proved to be of no use for quantitative interpretation because of the un- known amount of shielding involved. All these spectra gave results that were far too low in comparison with the other spectra. Since the "ac- tivity depressions' near these boxes all had about the same shape, it was decided that results from spectra recorded above the heater boxes were biased because of the boxes and not because of different deposition of fis- sion products in the heat exchanger. These spectra were discarded. Let us now turn to the different nuclides identified. Niobium-95 (Fig, 7.1). The disintegration rate along the longitud- inal axis ranges approximately between 0.10 and 0.30 x 10'® dis/min per square centimeter of heat exchanger tube. There appears to be an increase of activity close to the baffle plates in the range of 0.25 to 0.30 x 10'* dis min~™* e¢m™?, Between the baffle plates, the average activity level is near 0.10 to 0.15 x 10'? dis min~> cm™®., Since the activity increase is ORNL-DWG 70- 9570 3 | T T N T LIL D 0.30 Ef2 n ol S | | | }‘— ' o x | *] . .rfil | ~_ w 0.20 Ef2 . o * ., " i o i ’ g = . . | o s o . . - 'I . - . - g 0.10 Ef2 1 o l E — ~ 5 | ] T | HTR. PLUG SPACER HTR. PLUG HTR. PLUG | FUEL SALT FLOW —= | L || FPig. T.1. Activity of ?°Nb at reactor shutdown on June 1, 1969, in the MSRE heat excharger. PEe— L N 6§ 60 more pronounced for those baffle plates against the upper side of the shell, one may conclude that this activity increase is rather localized around the low flow rate areas near the shell and the baffle. There does not appear to be any general change in activity along the length of the heat exchanger. Ruthenium-103 (Fig. 7 2). The disintegration rate (dis min=* cm™? along the longitudinal axis follows nearly the same pattern as that for °SNb: along longitudinal axis, 0.17 to 0.65 x 10*'; near baffle plates, 0,50 to 0.65 x 10°'; between baffle plates, 0.17 to 0.24 x 10'*, Ruthenium—Rhodium-106 (Fig, 7.3). Ruthenium-106 is basically identi- fied by its short half-life decay product, *°°Rh. The activity distribution (dis min~’ cm™®) is essentially the same as for *°’Ru: along longitudinal axis, 0.04 to 0.14 x 10°*; near baffle plates, 0.11 to 0.14 x 10*'; between baffle plates, 0.04 to 0.06 x 10°*. Antimony~125. This nuclide was detected in small quantities all along the heat exchanger. The activity level, probably because of the very low fission yield, was small; the computer program only sporadically found peaks that could be assigned to '?°Sb, Variation of the activity level along the longitudinal axis was 0.053 to 0.14 x 10%° dis min~* cm™?2. There were not enough data to make other observations concerning the dis- tribution of the deposition. Antimony-126 and Antimony-127. These nuclides were both identified a few times in the different spectra. However, it does not appear sound to assign a quantitative value to those peaks in view of the very low activ- ity left after such a long decay time. Tellurium-129m (Fig, 7.4). The fact that both the fission yield of "29"MTa and the abundance of its photopeaks are small makes it difficult to detect this isctope with good precision (see also Sect. 4.3). There ap- pears to be no strong tendency of an activity increase near the baffle plates; however, a slight increase in activity near the cold end of the heat exchanger seems apparent. Variation of activity along the longitudi- 2 nal axis is 0.090 to 0.26 x 10'* dis min=' cm™2. Iodine~131 and Tellurium-Iodine-132. Most of the '3®'I and '?2*Te-I had decayed, although these nuclides could be positively identified in several spectra. The average activity of ***I at shutdown was approxi- mately 0.2 x 10'* dis min~* cm~*. The average activity of *®2?Te~I at -cm? OF HEAT EXCHANGER TUBE dis / min ORNL~ DWG 70-9571 | i T 060 El b t e e I e - e 1 . - S | e | | . ! I I 0.40 E .o e J—.——f * ; * | - ‘l | - . i . | ‘l | o I l . | . 00 @ 7 - 1 o 0.20 EN [— — — | . i - : . | | | \ | | | i | | | | | | | | | HTR. PLUG HTR. PLUG : SPACER HTR. PLUG l SPACER | 1 | | FUEL SALT E-I_LOW — Fig. T7.2. Activity of 1033y at reactor shutdown on June 1, 1969, in the MSRE heat exchanger. N N, , 19 EE ORNL-DWG 70-9572 Ll % [ : | B T C T 5 140 E10 ¢ ! i ol ‘ | . { | : . ‘ S : ! . | . | | o . ~ E 100 E1O | . | : . | 1 | . | E (5] e . . . O = * ‘ ! . | - {. .“ | - <1 o o® . | _ T 0.60 E10 o | - | . . I, , £ R | i | | . i e | ~ ; ! i | _— [ ! | w | ! . B | | | | | | | | HTR. PLUG HTR. PLUG ’ SPACER HTR. PLUG | SPACER ' | i ; | ' Nl J | | i ; : FUEL SALT FLOW | | s ' - ‘ | - | | | 1] Fig. T7.3. Activity of 106Ru~Rh at reactor shutdown on June 1, 1969, in the MSRE heat, exchanger, 29 dis /min * cm2 OF HEAT ORNL- DWG 70-9573 EXCHANGER TUBE ' T T l 0.30 EM : i , T o T - Ll * ® T I 0.20 Ef i S - A | P L | b 0.10 E# "o . .- L | t - r { f | | : | | | HTR. PLUG |~|+ HTR. PLUG E E | | FUEL SALT FlLOW—- Fig. T.L4. Activity cf '*°MTe at reactor shutdown on June 1, 1969, in the MSRE heat exchsanger. £9 64 shutdown was approximately 0.4 x 10'? dis min~' em~?. Taking into account the relatively short half-life of these isotopes in comparison to the decay time, not much precision can be expected of these values, 7.1.2 Main reactor off-gas line The flexible 1-in.-ID corrugated tubing, called the jumper line, was scanned, This 2-ft-long jumper line connects the main off-gas outlet at the fuel pump bowl with the 4-in.-diam section of the main off-gas system. This is about the only part of the off-gas system that could be "seen" from the portable maintenance shield and hence was the only piece surveyed. The 20 spectra were taken about six weeks after reactor shutdown (shut- down of June 1, 1969) at approximately l-in. intervals along the jumper 137 Cs line, Although there was some scatter in the data, especially for and *“°Ba-La, we could not detect a definite trend of decreasing activity in the downstream direction along the line. Although cesium, barium, and lanthanum were detected, these apparently were deposited in the off-gas line by the decaying xenon stripped from the salt rather than from salt deposits. Actually, no photopeaks could be positively identified from a nuclide that would indicate the presence of fuel salt; for example, °°Zr could not be positively identified\* Table 7.1 lists the nuclides that were identified; the activity is the average of the 20 different spectra. These data are extrapolated to the moment of shutdown (June 1, 1969) and represent disintegrations per minute per inch of off-gas line. It should be noted, however, that the main off-gas line was partially plugged during the latter part of the power operation prior to reactor shutdown. Off- gases were then routed through the fuel-pump overflow-tank off-gas line. 7.1.3 Main fuel lines Only a few spectra were taken from the main fuel line (line 102). Given the different counting geometry, due to the distance to the detector and the shape of the source, these results could only be qualitative. However, based on experience of the calibration experiments, we can expect % The minimum activity at which a peak could be identified in this case was «0.10 x 10°° dis/min per inch of line. 65 Table 7.1. Nuclide activities found in the jumper line of the main reactor off-gas system at shutdown on June 1, 1969 Activity Standard deviation Nuclide (dis min-? in.-1') of the average value (dis min~—* in.” %) *3Nb 0.22 x 10*?® 1.0 x 10** 1032Ru 0.45 x 10*3 2.8 x 10*? 1¢€Ru-Rh 0.19 x 10*? 0.8 x 10**? 123gp 0.36 x 10*? 5.0 x 10° 128MTe 0.55 x 10%? 5.4 x 10*° 1317 0.95 x 10** 6.2 x 10*° 137Cs 0.53 x 10*2 11.0 x 10*° 14%Ba~La 0.44 x 10*3 5.7 x 10**t 66 that the detection system has a fairly constant efficiency in the energy range 500 to 1000 keV. This would enable us to compare the activities of those nuclides that emit photons in that energy range. All nuclide activi- ties have been made relative to ?°Nb, because this isotope had the most prominent photopeak in the spectra at 765.8 keV. The nuclides identified, along with their relative activities, are given in Table 7.2. Restraint should be applied in using these numbers! It is intended to use these ratios only for comparison with the nuclide activities found in the heat exchanger and fuel line spectra taken in November 1969, Table 7.2 Relative nuclide activity found in the main fuel line (102) at shutdown time on June 1, 1969 Photopeak energy used in Activity relative Nuclide activity calculation (keV) to °°Nb >3Nb 765.8 1.0 193Ru 496.9 0.7 *°€Ru~Rh a 0.1 1238h 427.9 0.02 127gh b 12°MTe 459.6 0.4 13171 364.5 1927e-T e 7.0 aAverage of 511.8 and 621.8. Activity too small for proper calculation. “Average of 667.7, 772.6, and 954.5. 67 7.2 Group B Spectra These spectra were recorded through shield plug holes on the main off- gas line and heat exchanger. As mentioned earlier, holes were drilled in the lower and upper shield plugs to permit the recording of gamma spectra from the main reactor off-gas line, the heat exchanger, and one of the drain tanks. There was one through-~hole over each of these components; the steel containment membrane between the lower and upper shield plugs was, of course, left in place. Because of this membrane, the aiming of the collimator had to be done by trial and error rather than with the laser jig. We tried to obtain the maximum reading on a particular location by moving the collimator and detector back and forth. 7.2.1 Main reactor off-gas line Most of these spectra were recorded with the reactor at some power level and under different reactor conditions. The variable reactor param- eters were: use of argon instead of helium as purge gas, variations in pump speed, and different reactor power levels. The purpose of these spec- tra was to try to find a relation between the different noble fission gas concentrations (kryptons and xenons) and the changing reactor parameters. Especially at the higher power levels, the radiation intensity from the off-gas line was very high, in the order of 1000 R/hr or more. Fair doses of fast and thermal neutrons were also detected. This necessitated the ex- tensive use of shielding materials, and even then some spectra contained too many overlapping peaks for proper analysis but were still useful for comparative purposes. The different noble fission gases could be identified in most spectra and their concentration calculated in a major part of the spectra. These data were examined in connection with a study of '>°Xe behavior in the re- actor.'’ However, the values were too scattered to permit detailed analysis. '77. R. Engel and R. C. Steffy, Xenon Behavior in the Molten-Salt Reactor Experiment, ORNL-TM-3464 (October 1971). 68 7.2.2 Heat exchanger These spectra were taken to determine the change in fission product deposition on the heat exchanger after flush salt was circulated through the primary system, Spectra were taken before and after the flush-salt operation. Although the analyzed spectra did show the presence of noble metals, the activity was so low that it must be assumed that the collimator was improperly aimed and that we analyzed the general background in the high-bay area. This was confirmed by the analyzed data; there was no change in the activity of the considered nuclides before, during, and after the flush-salt circulation. 7.3. Group D Spectra These spectra were recorded through the shield-plug hole above the main reactor off-gas line during a beryllium addition. The purpose was to evaluate the possible change in noble fission gas concentration during and after a beryllium addition, but, as with the group B spectra, the scatter in the results precluded the determination of small changes, 7.4 Group E Spectra Just upstream of where the main off-gas line ties into the charcoal beds, there is a facility for isolating samples of reactor off-gas in any of three sample bombs. A hole was drilled in the shielding over the center sample bomb, and the off-gas flow was temporarily diverted and forced to flow through this bomb. Although no quantitative results can be expected, it is of interest to determine the ratio of the different fission gas con- centrations. Heavy shielding was necessary to cope with the high radiation level, As might be expected, some of the decay products of those identi- fied noble gases were also found. Because of the temporary diversion of the main off-gas flow, the quantitative activity value of these decay prod- ucts does not bear much importance, 69 Table 7.3 gives the ratio of activities of the identified noble fis- sion gases in relation to the ®°°Kr activity. It should be noted that the values of ***"Xe and '®®Xe are to be considered with less confidence. There was considerable scatter in the value of *®*Xe: +50%; the °°®Xe might be off as much as *60%. Table 7.3. Ratio of activities of identified noble fission gases in relation to the °®Kr activity Activity relative Nuclide to ®°Kr 87Kr 0.8 ®8Kr 1.0 ®°Kr Identified only 135Mya 0.05 135%e 0.6 13%%e 0.07 7.5 Group F Spectra These spectra were recorded through the shield plug holes of the heat exchanger, the main off-gas line, and the drain tank. From previous ex- perience we knew that it would take at least two days to remove the upper shield plugs from the reactor cell, cut the containment membrane, and set up the portable maintenance shield (PMS). The holes in the shield plug would give us a convenient opportunity to record spectra at those locations during the time-span between reactor shutdown and installation of the equipment on top of the PMS. Drain tank data were taken through the hole only. 70 There was some disagreement between the data recorded through the holes from the heat exchanger and off-gas line compared with the data taken at the same locations with the PMS installed. It seemed that the disagree- ments were due to misalignment of the equipment, and the results were ad- justed accordingly by normalizing to a known nuclide (°°Nb). Several nuclides that otherwise would have been totally decayed could be identified and their activities estimated. The detector and collimator were moved several times during this two-day period from one location to the other. Reactor shutdown and drain from full power occurred on November 2, 1969, at 1441 hr. The procedure was as follows: with the reactor still at full power, the thawing of the system drain valve was requested. Once this freeze valve was thawed, the drain of the fuel salt caused a drop of the fuel=-salt level in the fuel pump bowl which automatically stopped the fuel pump and then scrammed the reactor. 7.5.1 Main reactor off-gas line The collimator and detector were set up over the off-gas line hole two days before the actual shutdown and drain of the system. Several spec~- tra were thus recorded with the reactor at full power; heavy shielding was necessary for the recording of these spectra. Because of the very large number of photopeaks in the spectra during and shortly after reactor operation, it is obvious that many minor peaks would go undetected in the analysis and sometimes would be added by the computer program to the areas under the larger peaks. This leads to a cer- tain overestimation of some of the nuclides present. TFor the evaluation we have tried to select those photopeaks that seem to be relatively iso- lated. Another check on the validity of a selected photopeak was the iso- tope half-life time as deduced from the presented graphs. Because of nuclide activity variations due to sometimes short-half- lived precursors, all these off~gas line data are presented as activities measured at the moment of counting time rather than extrapolated back to reactor shutdown time. 71 First, the spectra were recorded with the reactor still at full power, and from these the average activity was calculated for the identified noble fission gases. Although many isotopes with longer half-lives were also identified, their activity appeared to be highly influenced by the many large photopeaks of the noble fission gases. An exception was °°Nb, with 1 a calculated average activity of 0.99 x 10'? dis min™* in.”", which is close to the "best estimate" of the ®°Nb activity of 0.90 x 10*? dis min~' in.”* given in Section 7.6.2; all activity results are normalized to this last °5Nb activity. Table 7.4 shows the activities of noble fission gases de- tected in the off-gas line with the reactor at full power; we estimate the uncertainty of these results to be *40%. The activities reported in Table 7.4 appear to be much too high. If one calculates the maximum nominal disintegration rate of these fission gases in this section of the off-gas line, the values in the table appear to be a factor of 10 to 50 too high. The nominal calculation takes into account the yield of the fission gases and the reactor power and allows no holdup time in the primary loop. We found no way to explain this dis- crepancy except to postulate a longer than normal residence time for these fission gases in the off-gas line §such as might be produced by adsorption on the wall or on other deposits that were known to be present). Let us now consider the activities of different nuclides after reactor shutdown. These spectra, being recorded through the shield plug hole, have also been normalized to the best estimate of the °°Nb activity. It appears that most photopeaks from spectra taken at 1.04 and 1.49 days after reactor shutdown yield results that are too low. Although the cause 1is unknown, little confidence should be placed in the results calcu- lated from these two spectra; these points are indicated in Figs. 7.6 to 7.20 as black points. Krypton-87 (Fig. 7.5). There is a rapid activity decrease at the start of the fuel-salt drain, subsequent fuel pump stop, and reactor shut- down. The raising of gas bubbles from the salt after the fuel pump stop, the release of fission gases from the graphite, and the possible back surge of decaying fission gases back into the system because of drain might all be contributory to the first rapid and then slower decrease of activity. 72 Table 7.4, Noble fission gas nuclide chains identified in the main reactor off-gas line with the reactor at full power Nuclide Average activity (dis min=* in.”? 87kr 0.13 x 10*® 88Ky 0.12 x 10*° ®®Rb 0.18 x 10** ®9Kr 0.22 x 10*%% #%Ru 0.82 x 10** °Okr 0.73 x 10** >°Nb 0.90 x 10'° 133%e 0.13 x 10** 1357y e 0.50 x 10** 138%e Identified only, order of activity: 10%* 13%%e Identified only, order of activity: 10%3 13%¢Cs Identified only, order of activity: 103 149%e Identified only, order of activity: 10%? 73 ORNL-DWG 70-9581 5 - - —— - —_— | v - e 4 4 f ‘ , | | AVERAGE 87kr ACTIVITY WITH THE REACTOR | | ] T POWER AT 8 MW 2 o , o 10M _ S S R R R R L I . o 1 __,__ I | ) o I _ . A 5 . — - 1‘ B 5 L. —~AVERAGE '°3Xe ACTIVITY WITH THE REACTOR . A POWER AT 8 MW ‘ ! fix \ | : I \ : 10"3 e — [, - ; s ] —_ - — e e e ———— Q e S — - _——— oo —t + £ - } - -t _— + - v oL T L ' = AT T T T o 8r: 2556.0 kev o - Ly e S S S . 135, N © A A A Xe, 249.6 key < : ; < . < 2 ‘fi £ | ["s] I : | a2 B ! w0 o1& S ST T ] - . O 2 = S < \ . e - . . e e e e i e e 5 b ; . ) 1 I i . ] ] e . ; ] | f 2 s e e *;777777 + - — A - - 10" e 4o o, el , e ——— e e —— . .- i + 4 - o e - 1 - . 4 i . . . 5 b e 4 - . . . . L . - o . . T § ! 2 . - 1 + + | | ; j : ; 1010 ; | i ! 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 DECAY TIME AFTER REACTOR SHUTDOWN {doys) o O n Fig., 7.5. Activity of 87Kr end '35Xe after reactor shutdown in the jumper line of MSRE main off-gas line (November shutdown). 74 Krypton-88—Rubidium-88 (Fig. 7.6). There is a similar rapid activity decrease with °®Kr; the °°Rb activity decreases almost at the same rate as 88, I\r- Strontium-91 (Fig. 7.7). This nuclide is a decay product of the °'Kr chain. The decrease in activity fits the decay half-life well, which would confirm its identification., Extrapolated back to zero time, one can esti- mate its maximum activity to be 0.20 x 10'® dis min™'! in.”'. This extrap- olation seems reasonable if one considers the short decay half-1life of its precursors. Niobium-97. This nuclide was identified only a few times directly after reactor shutdown. Its activity was in the order of 10'* dis min™* . -1 ln. - Molybdenum-99 (Fig. 7.6). Although a few of the data appear somewhat too low, most of the activity results decrease according to the °°Mo decay half-l1ife. From the actual survey of the jumper line, we estimated the °°Mo at the moment of shutdown to be 0.33 x 10'®*, The activities plotted in the figure agree well with this estimate. Ruthenium-105—Rhodium-105 (Fig. 7.8). The calculated activity of 195Ry is deduced from two of its photopeaks; these agree well. The decay of the activity conforms very well to the literature data for its decay half-life. Extrapolated back to reactor shutdown time, the estimated maxi- mum '°’Ru activity is 0.34 x 10%? dis min~' in.”%. The daughter nuclide, *°°Rh, is rather difficult to identify because its photopeaks are not very prominent and are close to peaks from other nuclides; one might expect erroneous results. The maximum '°5Rh activity is approximately 0.8 x 10*° dis min~* in.”'., The amount of '°°Rh present in the off-gas line could be due to the buildup from the decaying *°°Ru as '°°Rh separation from the salt. well as from the separate mechanism of Antimony-129—Tellurium-129m (Fig. 7.9). In a few spectra we could identify '2°Sb; the lack of dependable nuclear data (decay scheme and abundance of its gamma ravs) made it impossible, however, to assign an ac- tivity to this nuclide. O—h o N — 1) ~o NUCL IDE [ (dis/min)/in. of of¢-gas line] ~ o ORNL-DWG 70-9582 ACTIVITY OF | Kr WITH REL\CTOR AT 8 MW I g e ; : + e C———— —t o 4 - e :, e : Lo . — — - + —_— -~ + 4 T ~ACTIVITY OF 88Rb WiTH REACTOR AT 8 MW - > o 1529.8 kev —— —— s G/ b 23920 kev ¢ 898.0 kev ¢ 1836.1 kev A 739.7 kev ! ; _ A 778.2 kev R Do — FE e e b L e 7T77 e - § e .,.—%—7,._~.a.._..._._g b e ; . | . b i 3 | i ! 0 0.2 0.4 Fig. T7.6. 0.6 0.8 1.0 DECAY TIME AFTER REACTOR SHUTDOWN Activity of 88Kr, MSRE main off-gas line (November shutdown). 2.0 2.2 2.4 2.6 2.8 3.0 (days) and ?°Mo in the Jumper line of the ORNL-DWG 70-9583 LT T T | - i et { | “W'W‘:;"”W' ] ] ‘ . R S | ',,44 | } | i Mgr. o 1024.2 kev | : ‘ 0 B847.0key |- At ¢ 13411 ¢ 884.1kev 4 1072.5 kev : [ e e R - S + . o - S e - S i e o - | b e e e - e . < o * g b boee e e b . go\q e e i “{, - . e e e e e v [ EEL - M k. 1, f § { o | | | | g o s, NUCLIDE [(dis/min)/in. of off-gas line f b | | | ; | 1010 l ) i 1 L I z | 0 0.2 0.4 0.6 0.8 1.0 .2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fig. 7.7. Activity of 2'Sr and '**I in the jumper line of the MSRE main off-gas line (November shutdown). 9. NUCLIDE [ (dis/min) /in. of off-gas line | 43 o ORNL-DWG 70-9584 1O g ——a - 4 T T _?)_g . .w : 1 - ifi*“— — : ] _ _ qo-0 L e ; _77;__»,,__,, _ 5 .- & & ] | 4 | | { _ o ! AH 5 e ' } —— e S — . 5 2 -Q- f t @ N+ ‘ | 1012 & ‘ : | ; ‘ B e , ‘ % e | - 7H¢ ?\L‘i : 105Rh 0 306.3 kev ‘ . 7_77 : 5 ¢ - & 319.2 kev ; | ; | (04694 kev - e * P e 109-y 14 676.3 kev - - e i | | | 2 4 724.2 kev 21 v T T | | | | 1 } o : | i ol o . 0.2 04 Q6 08 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.9 2.6 2.8 30 Fig. 7.8. DECAY TIME AFTER REACTOR SHUTDOWN (days) Activity of 195Rh ana !'°°Ru in the Jumper line of the MSRE main off-gas line (November shutdown). LL of off-gas line] NUCLIDE [(dis/min}/in. s ORNL -DWG 70-9585 10 T T T 1 : T , . SR ok . Lo - '.,*,, el ,,,,,,%,.,, i e . {_ - I . e e z 429Te; 0 459.6 kev R ! b - e -4 i . e S — e 'f .- T — 5 - e e e - e e e = — - . --4‘—----— R - o I 0 793.8 kev ‘ | ' T 13mredd 852.3 kev ’j T Ty 4 1206.6 kev - - O i E ‘ i \O " ' ! .o : 0% L ” o ; i ol e e e e e P e i - ] s| 0 ; ¥ I | e e : ; — . . v | | . S | 0 s ol b I R | i s ‘ ! ! ! 1ot j e } | ; 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fi .o 129 1317y, . ig. 7.9. Activity of Te and e after reactor shutdown in the jumper line of the MSRE main off-gas line (November shutdown). 3.0 8L 79 Tellurium-129m was detected by its decay from the isomeric state to the ground state and the subsequent disintegration of the ground state. We identified '?°"Te primarily by the most abundant photopeak of the '#°Te daughter at 459.6 keV. In the decay of *?°8b, the branching ratio is 85% to the ground state, 12°Te, and 15% to the metastable state, **°"Te, Since the half-life of 129Ta jitself is short, the decay of its 459.6-keV photopeak should, after a short time, follow a composite of the decay half-lives of 12%5b and '*°MTe, Figure 7.9 shows the activity related to '*°Te. The low yield of only 7.7% for the 459.6-keV photopeak in the decay of '?°Te implies also that any experimental scatter is amplified by more than a factor of 13 for the activity calculation. This would, of course, explain the state of the results. The '2?°MTe activity can be deduced from Fig. 7.9; this activity ex=- trapolated to zero time agrees reasonably with the estimate of Section 7.6.1. The initial faster decrease of activity of **°Te, with a half-life 129 of about 5 hr, could possibly be attributed to the Sb decay. Tellurium-131m (Fig. 7.9). Based on the results from three *?'MTe photopeaks, the activity of this nuclide could be well established. Its identification was confirmed by the decay half-life as deduced from Fig. 7.9. Extrapolated back to reactor shutdown time, the estimated maximum activity of *?*Te is 0.75 x 10*? dis min~' in.~'. The maximum activity calculated by extrapolation to reactor shutdown time is acceptable since one can discount the effects of an eventual ®?Sb decay. Only a few photo- peaks in the different spectra could be identified as '317e; these results were too sporadic for any interpretation. Todine-131 (Fig. 7.10). With the exception of two sets of data (1.04 and 1.49 days after shutdown), the results are in good agreement. It ap- pears that the maximum activity occurs shortly after reactor shutdown time. This maximum activity is 0.5 x 10*® dis min™' in.”', somewhat lower than the best estimate (Sect. 7.6.1) of the jumper line survey. It might be ex- 13MMTa, Another argument is that plained by the influence of the decaying 1317 pight “evaporate" or transfer from the hot, empty fuel system and set- tle in the off-gas line (see Chap. 8). NUCLIDE [(dis/min)/in. of off—qgas line] 1012 ORNL—DWG 70— 9586 . 1 { 1 z I [ T | : . : ‘ | J \ 5 ; ; 131, [© 364.5 kev " o * L j ‘ } ®637.0 kev —— ! ) o QO g o 8 | 1 . . i ' % g e e o ‘ . | | | ] ¢ W | | | % ‘ | B K | | | T ‘ i | , | ! } ! ‘ . [ : ' | i | | | | | 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.5 2.8 3.0 DECAY TIME AFTER REACTOR SHUTDOWN {days) Fig. T7.10. Activity of 1317 gfter reactor shutdown in the Jumper line of the MSRE main off-gas line (November shutdown). 08 81 Tellurium-132—TIodine-132 (Fig. 7.11). There is considerable scatter of both the !32*Te and '°2?I results in the first hours after reactor shut- down. During the first 10 hr, disintegration rates vary between 0.45 x 10*° 132 and 1.3 x 10*® dis min~' in.—'., Since three I photopeaks seem to yield consistent activities among each other, one has to accept apparently an 132 f extra increase o I activity after shutdown. This cannot be explained 132 132 from the decay of Te and I alone. This surge in activity could be 132 conveniently explained if it is assumed that some of the I formed by decay of '32Te in the hot, empty fuel system transfers to the off-gas line along with the circulating purge-gas flow. An extrapolation to reactor shutdown time, based on the data taken after 2.3 days and taking into ac- count an equilibrium condition between '*2Te and '®2%I, yields 0.67 x 10*® dis min=* in.~', in agreement with the estimate in Sect., 7.6.1 (0.65 x 10*®), However, if one admits a transfer of iodine to the off-gas line, a maximum *?*I activity of 0.12 x 10*“ dis min~' in.”' can be expected. Todine-133 (Fig. 7.12). The results are in reasonable agreement with the *?°I decay half-life. The activities calculated during the first 6 hr f 133 seem to be low, which might again indicate that some transfer o I occurs from the reactor system to the off-gas line after reactor shutdown time. Taking into account an equilibrium condition of Te-I, and the possible 133 iodine transfer, it is estimated that the maximum I activity after re- actor shutdown is 0.17 x 10*® dis min-! in.™?. Todine-134 (Fig. 7.7). The later spectra yield activities that are in very good agreement with the expected decay half-life value. It is esti- 1347 activity after reactor shutdown is in the order mated that the maximum of 0.30 to 0.40 x 10'3® dis min~?! in.”*. Xenon-135 (Fig. 7.5). Again there is a rapid decrease in activity di- rectly after reactor shutdown which might be explained by the different circumstances related to the stoppage of the fuel pump and the drain. Even several hours after reactor shutdown and drain, the '®°Xe activity is still appreciable, NUCLIDE [(dis/min)/in.of off-gas line] ORNL-DWG 70-9587 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fig. T.11. J | o 667.7 kev i i _ | 321 J4 9546 kev 1 J ;{> $ gg ! 4 . #4398.6 kev . 3 R S 8-- —£~~~r : 4 . 3271 52282 kev | T ? g___ 00 o O + , .o :; i+ e 4{. e ' | . - - . o ; . . . | ;?1? ; t 'I' b e o ¢ , ; . ‘ | | i : i . | x | A j 1 oo : : ? f J b - : + : - - + + - l -— ! | 1 | | i i | | | ! ] ' ' 1 L [ | i 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Activity of 1327e and '3%I measured after reactor shutdown at the jumper line of the MSRE off-gas line (November shutdown). c8 NUCLIDE [(dis/min )/in. of off-gas line] 0.2 0.9 0.6 Fig. T.12. ORNL-DWG 70-~9588 —————— L + 1331 6 529.9 kev 1 | 0.8 .0 .2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 DECAY TIME AFTER REACTOR SHUTDOWN ({days) Activity of 1331 after reactor shutdown in the Jumper line of the MSRE main off-gas line (November shutdown). £8 84 7.5.2 Heat exchanger With the reactor at full power, the radiation coming through the shield plug hole over the heat exchanger was too high to record any spec- tra. The main problem was that the setup of the collimator with detector caused such a high dose rate of scattered radiation during the alignment procedure that it was judged unwise to continue as long as the reactor was at power. Had the collimator and detector been set up and aligned before the reactor was at power, we could have recorded spectra; however, we found in later experiments that spectra taken over the heat exchanger with the reactor at full power could not be analyzed anvhow because of too many peaks. This situation also prevented the recording of heat exchanger spectra until the collimator was properly aligned after the drain. Thus we lost almost three wvaluable hours of recording. With our location equipment we determined the position of the shield plug hole and hence could relate the calculated activities found from the spectra through the shield plug hole with those deduced from the actual sur- vey afterward. All calculated activities were normalized by comparing the ®31b calculated from a spectrum taken through the shield plug hole and that actually found during the survey (Sect. 7.6.2). Ve used 0.93 x 10** dis min~' em™® as the normalizing activity for °°Nb at this location. All ac- tivities given below are calculated at counting time and are expressed in disintegrations per minute per square centimeter of heat exchanger tube surface. In view of our calibration methods, it should be noted that the ac- tivities of gaseous fission products, which are obviously in the space in the shell side of the heat exchanger, are expressed in equivalent disinte- grations per minute per square centimeter of heat exchanger tube. Krypton-88—Kkubidium-88 (Fig. 7.13). Even 3 hr after reactor shutdown, the ®°%Kr and °° Rb activities are still appreciable. Comparing the krypton and rubidium activities, it appears that these nuclides are close to an equilibrium condition, Strontium-91. A few spectra indicate the presence of °*Sr, with its photopeak at 1024.,3 keV. Since we found ®®Kr, one might, by the same token, expect decay products of °*Kr. The data were too sporadic to really indi- cate an average activity; 10 hr after reactor shutdown, a few spectra 10” ORNL-DWG 70-9589 T e e S T et 10’0 b e N o © NUCLIDE 2 - O S | | (dis/min)/:m2 of heot\exchanger 1ube] v a-o%0- e e e . . 88 0 898.0 kev § 1836.1 kev — A : '- I S P ] + § . e . e e e o - 1_ o - b ] t | . o . - . + + . i . y . . - ‘ - e 7,vt_,777 . ‘ ‘ . ‘ ' . e e — . { ‘ b ' ' : ! . + e i 7;, C e ,.4:_, bl ; : | ' Ly T ? r , T ; b = | | i | ; | P ' . - e ——«‘——w»_ 4 e . e | 1 , ; | e ] ey i + ¥ 1L _ i — — - ' : | , ‘ 1 O 8347 kev ! ‘ — 88K 16 1529.8 kev - - ‘ _ - 4 2392.0 kev | e e 4 e b ; 4 e i e s + 108 | 0.4 0.6 08 1.0 1.2 t.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fig. 7.13. Activity of ?®Kr and ®®Rb after reactor shutdown in the MSRE primary heat exchanger (November shutdown). ¢8 86 indicated a ®'Sr activity of approximately 0.15 x 10*° dis min~' em™? {not extrapolated back to zero time). Niobium=-97 (Fig. 7.14). This nuclide has only one strong photopeak (658 keV), which makes it hard to assign much confidence to its identity, especially since many other peaks in these spectra are located very close to it. The fact that the observed photopeak decayed with a half-life of 108 min indicates there was interference in measuring the peak. We believe, however, that this decay rate is sufficiently close to the accepted half- life (78 min) to justify the peak's assignment to ®’Nb. Extrapolation to reactor shutdown time is acceptable, since its pre- cursor (°7Zr) was not identified in the heat exchanger. The estimated maxi- mum activity at zero time is 0.6 x 10" dis min™' em™ 2. Molydenum=-99 (Fig., 7.14). Judging from the plots of different spectra, we have more confidence in the activities yielded by the 739.7-keV photo- peak than in those from the 778.2-keV peak. The latter is located very close to a large **°I and °°Nb peak; this would make the activity calcu- lation somewhat doubtful, especially since other short-lived isotopes also have peaks in that energy range. The activities yielded by the 739.7-keV peak are rather consistent and in good agreement with the results of the actual survey afterward. The estimated maximum activity at reactor shutdown time is 0.18 x 10*? dis - o 1 -— min cm T . Ruthenium-105—Rhodium-105 (Fig. 7.15). It appears difficult to evalu- ate exactly what the activities of these two nuclides are. One would expect that *°° Ru and *°?Rh behave approximately the same. The three '°°Ru photopeaks (469 4, 676.3, and 724.2 keV) decay with a half-life close to that of '°°’Ru. The first two peaks, however, yield a nuclide activity almost half that from the third peak. The activity de- duced from the first two peaks at reactor shutdown time would be 0.3 x 10%* dis min~! cm~?; that from the 724.2-keV photopeak would be 0.46 x 10" dis min~* cm™®. It is very well possible that our nuclear data are not correct for this isotope. '°5Rh activity results is even worse. Especially The scatter of the during the first hours after shutdown, the 319.2-keV peak indicates a high activity; these possible aberrations might be due to photopeaks from other nuclides, which could cause this overestimate. NUCLIDE (dis/min)/cm2 of heat exchonger tube] [ S, o ORNL—DWG 70-9590 & | | — | ' 1 : ‘ ‘ ‘ : ol L ee, ot . | * ‘ | \ | Vo778 2kev ; | . | | | | °"Nb 0 658.2 kev ! ey e % e e = T L + - —_— ;{ e R r— e ,’__;_-‘—— — ~L — + - } — - + t —————————fr———— <|+-—~ ————— —_ i —— — I U SN f - — & . G e 3 _ —_—— TN ek + - , - ‘ f ; et e 0 i ‘ i : , | T — ] | "* - et e w-qt-m e e ] - + _—t —_ — —o l ‘ L e —— ‘ - | t 0.2 04 0.6 Q.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 30 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fig. 7.1k, Activity of ’Nb and °°Mo after reactor shutdown in the MSRE primary heat exchanger (November shutdown). L8 NUCLIDE [(dis/min}/cm? of heat exchanger tube] ORNL—DWG 70—259f o' — T 1 T T T | 1 ' . . T ’ i _ ‘ -405 IO 3063! keViii:‘ihuj‘..:'.—_“ ] [ | | ~ ‘ * - +— 6 319.2 kev —+ ¢ z | - - 0 469.4 kev ¢ o | | | I | +{O5Ru{¢ 676.3 kev ¢+ 1 | , ; l 3 © B ¥ 724.2 kev I _Q_ : | ‘ ¢ , : y 8 4 i o | Q o 100 o 7 | | : N . 1 L | | r l [ oy S T __Q_f__ - 1 8 | | |o | 1 _1 T ~ 5 o . 4 . i 4 + T ] - R — M+ e L T”, 1 L ] ol o % S N S R 00 | | | | | 02 04 06 08 0 42 14 16 48 20 22 24 26 28 30 Fig. 7.15. DECAY TIME AFTER REACTOR SHUTDOWN (days) Activity of '95Rh and '°SRu after reactor shutdown in the MSRE primary heat exchanger (November shutdown). 88 89 Based on activities deduced from later spectra, we might conclude the 105 maximum Rh activity would be in the range of 0.4 x 10! dis/cm?®. Antimony-129—Tellurium-129—Tellurium-129m (Fig. 7.16). Since several other antimony isotopes are known to be present on the metal surfaces of the heat exchanger, there is no apparent reason why ‘2°Sb should not be present too. The problem is that the decay scheme of this nuclide is not known with certainty; some of its gamma rays are indicated in the litera- ture, but their reported abundances are doubtful. We believe we have identified '2°Sb by one of its photopeaks (1028 keV) and by its decay half-life; but, due to lack of data on the branching ratio for this gamma ray, we did not calculate the disintegration rate of *2°Sb. The photon emission rate of this gamma ray at reactor shutdown time was 0.90 x 10*° photons min~* cm™2, As reported in Section 7.5.1, we identified '?°™Te primarily by the 459.6-keV peak from the decay of *2°Te. The intensity of the 459.6-keV (and 1084.0-keV) peak would reflect the decay of **°Sb, *?°Te, and '**"Te, Figure 7.16 shows the '?°Te activity based on the 459.6- and 1084,0-keV 129Te photopeaks. The absolute abundances of these peaks per decay of (ground state) are taken as 7.7 and 0.6% respectively. In the same figure is shown the photon emission rate of *2°Sb for its 1028-keV peak. Tellurium-131m (Fig. 7.17). Even if '?'Sb would deposit on metal sur- faces, the maximum *>'"Te activity would occur, because of the short *°'Sb half-1life, very soon after reactor shutdown time; hence an extrapolation to zero time for the maximum activity seems justified. The 1206.6-keV photopeak appears the least influenced by other peaks and should be con- sidered the most trustworthy., Our estimate of the maximum *°'"Te activity is 0.73 x 10** dis min~' cm™?, Todine-131 (Fig. 7.18). Based on the chemical properties of iodine, there is ample reason to believe that iodine remains with the fuel salt. This was confirmed by gamma-ray spectra taken in a later stage of the ex- periment. This would mean that any iodine detected came from the decay of either antimony or tellurium, The half-lives of '®'Sb and '*'Te are so short that the buildup of 1317 from these nuclides could not be observed. The formation of iodine ¢ [photon emission rc:'re/cm2 of heat exchanger tube ] ¢ NUCLIDE [(dis/min)/cm? of heat exchanger tube] ORNL-DWG 70—-9592R | | - | | ‘ 10" -0 ) - [‘ - *fif"f___;fl*:_ - ‘Jf“ I““fiL‘ i:j ey o = o | j_m = 5 | el . o ' ! L —[ ‘Jr_ IL ; i —_ Oy T _— + - e — { 4" e 4 g e - ! w-————m——+—i —A—] | I L - _ S 3 | b | o 2 |— _ [ o N . i -5 e q | .[ ] l ~ o | 10'% | ¢ P : o ° ) LT - B T e T T 'fi;gfig““”ii:;m:”fif” "ff - ~ N L T 5 __._o__", i — ! e L 129T 0 459 .6 kev 1 - ' 16 1084.0 kev oy e b "%b ot028kev — i . b e | 0 | . i 10° N | j - T T T 8T _ ) — e 4 e VV—A—-T————Af - - ,‘ —_—— 5 : | 1 . ‘ 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 {.8 2.0 2.2 2.4 2.6 2.8 3.0 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fig. 7.16. Activity of '?°Te and photon emission rate of '??Sb mea- sured after reactor shutdown in the MSRE primary heat exchanger (November shutdown). 06 91 ORNL—DWG 70-9593 10“ - 1 ‘3t © 793.8 kev Te 4 § 852.3 kev 1206.6 kev 847.0 kev 134 $884.| kev 4~ 1072.5 kev NUCLIDE [(dis;/min)/Cm2 of heat exchanger tube] 0 0.2 04 0.6 0.8 1.0 t.2 1.4 1.6 i8 2.0 2.2 24 2.6 2.8 3.0 DECAY TiME AFTER REACTOR SHUTDOWN {days) Fig. T.17. Activity of 131mpe and !3*I after reactor shutdown in the MSRE primary heat exchanger (November shutdown). ] N - ol = N (dis/min)/em?2 of heat exchanger tube o NUCLIDE [ 3, o ORNL-DWG 70-9594 T T T T ! o | I ! 0 364.5 kev : L . , . R 134 _ b . e {¢ 637.0 kev ¥ , s r s - mfmn . ; 1 e e “ » ; e - . } | : ; ¢ 1 g . L o ) j¢ ¢ - N | ¢ . . 0 . R 0? \ ’ ! ? 0 ¢ ! 0 o ! 3 09 ° | | 1 ; i i 0 0.2 0.4 0.6 0.8 {.0 1.2 1.4 {.6 1.8 2.0 2.2 2.4 2.6 2.8 Fig. T7.18. mary heat exchanger (November shutdown). DECAY TIME AFTER REACTOR SHUTDOWN (days) Activity of 1317 after reactor shutdown in the MSRE pri- 3.0 4 93 from the decay of '®'"Te, in the case that primarily *°'Sb deposits on the metal surfaces, presents only a few percent of the total iodine activity and is difficult to observe. A simple extrapolation to reactor shutdown time to obtain the maximum *3* I activity would then mean a very small over- estimation because of this decay of *3*'Te, In case tellurium rather than antimony deposits, the overestimation would be much higher, in the order of 60%, because of the slow decay of 131/MTg, There is a fair degree of scatter in the calculated activities. The estimated maximum activity of *®'I would be approximately 0.24 x 10'' dis min=' cm~? and is based on the average of the 364.5- and 636.9-keV photo- peaks. This conforms with the results of the actual survey by extrapo- lating those back to reactor shutdown time. The evolution of the iodine activity is considered in more detail in Chapter 8. Tellurium-132—Iodine-132 (Fig. 7.19). For the calculation of the '32Te activity, a simple extrapolation to reactor shutdown is justi- maximum fied. Even if it is the tellurium precursor, **2Sb, that plates out on the metal surfaces, the latter decay half-life is so short that this would jus- 132 tify the assumption. The maximum Te activity is approximately 0.40 x 10'? dis min™' cm—2. The iodine activity is expected to grow in from the tellurium decay; that is, its maximum activity should occur about 9 to 10 hr after reactor shutdown. The observed *32?I activities do not agree with this. Its ac- tivity went down again after 5 hr and then after 20 hr or more finally came into equilibrium with its precursor. The excess of **?I in the main off- gas line at about the same time period might indeed point to the transfer of iodine from the reactor system to the off-gas line. Because of this temporary deficiency in the heat exchanger, the maximum **?I activity occurs -2 some 24 hr after shutdown time and amounts to 0.34 x 10'? dis min—! em™2. Iodine-133 (Fig. 7.20). The *3?I activity seems to decrease faster than can be explained from decay only (transfer!). Taking into account 133 the decay of its precursor, the maximum I might be expected to be below 0.20 x 10'! dis min=* cm™® (before its apparent transfer). U M 1042 N NUCLIDE [(dis/r'nin)%m2 of heot exchanger tube:’ o — —_ — 'ORNL-DWG 70-9595 Fig. T7.19. DECAY TIME AFTER REACTOR SHUTDOWN (days) MSRE primary heat exchanger (November shutdown). ‘ ;' | ; J[ R l L : ‘, | o - — ;‘] ; ; . l ’ ; L I e . > 3 1321¢ { o 228.2 kev f b ; | o ____’%_ . I | g 0 667.7 kev r ; | | 321 1§ 954.6 kev | | I | 4 1398.6 kev 1’ . 3; : : ’ ) S T T | . l . § , } e . +‘ e e | . % . . . S 4 e g fi L o . i . o4 .l L T ] 0 o o % : i i +§ oy | © o O $ 4 | L T_ ,g‘¢, | “1} RS e v | | ° © | : l i ! ‘ | | | ; ‘ | | R o ; 1 | | 0 0.2 04 06 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Activity of 1327e and '32I after reactor shutdown in the %6 NUCLIDE [(dis/min)/crn2 of heat exchanger tube] ORNL-DWG 70-9596 35%e: © 249.5 kev _ ___ 1331, 0 529.9 kev 0.2 04 0.6 Fig. T.20. ¢ 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 DECAY TIME AFTER REACTOR SHUTDOWN (days) Activity of 1337 and !3°Xe after reactor shutdown in the MSRE primary heat exchanger (November shutdown). ) 96 Todine-134 (Fig., 7.17). The maximum '°*I activity might be expected toc be around 0.30 to 0.60 x 10'! dis min=! cm~? if one takes into account 134 that I 1s on the metal walls only as a decay product of tellurium, Xenon-135 (Fig, 7.20). The presence of xenon in the reactor system is ample and in the first hours after reactor shutdown forms an important ac=- tivity source. Its maximum activity might, at that time, run as high as 0.30 to 0.40 x 10'* dis min=" cm™?, The decrease in activity is obviously due to decay and purging of the system. 7.5.3 Drain tank Several spectra were taken from one drain tank after reactor shutdown, These spectra were recorded through the hole in the shield plugs over the drain tank. There was no possibility for calibration of the results or even for proper alignment. By moving the detector in small amounts back and forth, we tried to obtain a maximum activity reading. All data, of course, had to be normalized to have a basis for compari- son. We chose the 724.2-keV photopeak of ®°Zr, extrapolated back to reactor shutdown time, This appeared to be the most convenient normalization point, since zirconium is supposed to remain with the salt and will be at its maxi~- mum activity very shortly after reactor shutdown time because of its short- half-life precursors. The spectra were taken primarily to check the evolution of the niobium activity, However, there were some other interesting observations too. Since we were not entirely sure about the location of the place on the drain tank from which spectra were taken or the amount of shielding ma~- terials (flanges, drain-tank vessel, steam drum, etc.), we will report only on relative results of nuclides having photopeaks close to each other in the energy range of 600 to 800 keV. Table 7.5 indicates the evolution of the ratio of the °°Nb to °°Zr and the ?’Nb to ®’Zr activity. The °°Zr activity is based on the average of the photopeaks at 724.2 and 756.9 keV. The °°Nb, °’Nb, and °7Zr activi- ties are based on the 765.8-, 658.2-, and 743.4-keV photopeaks respectively. The °’Nb/°’Zr ratio is not of great value for analysis, since any °’Nb iden- tified is formed in the drain tank in our case. (Any °’Nb in the salt that 97 Table 7.5. Ratio of ®°Nb to °°Zr and ?'Nb to °7Zr activity in relation to decay time after reactor shutdown in November Activities at counting time were used; no extrapolation to reactor shutdown time Decay time after reactor shutdown (days) ®SNb/®%zZr *’Nb/®7Zr 0.872 0.78 0.97 0.971 0.79 0.99 1.982 0.61 0.97 2.797 0.62 1.02 2.873 0.65 3.920 0.97 4,900 1.55 7.780 2.07 8.893 1.98 11.904 2.30 11.963 2.30 15.885 2.65 16.907 2.89 98 drained from the reactor would have decayed before these spectra were taken.) It is, however, a good check on the credibility of the °°Nb/’°Zr data. The °°Nb/°®Zr ratios do not appear to indicate the real activity ratio °5Nb in the reactor in the salt itself. If we consider all of the °°Zr and system (regardless of location), the maximum value for this activity ratio is 1,0, immediately after shutdown. In view of the ?°Nb deposits in the primary loop, the activity ratio in the drain tank would be expected to be appreciably smaller, unless the detection efficiency for ?3Nb were somehow enhanced. One explanation might be that the °°Nb concentration near the salt surface is much higher than the average in the salt. The first spectra taken after reactor shutdown (after about 0.800 day of decay) revealed the strongest activities to be due to °°Kr(Rb) and '*°Xe. Besides the nuclides expected to be with the salt, such as yttrium, zir- conium, cesium, cerium, and barium-lanthanum, there were also some unex- pected observations. For example, we did not detect any molybdenum, anti- mony, or tellurium; the amount of ruthenium present (both *93Ru and *°°Ru) was very small, Molybdenum-99 has several photopeaks by which it could be identified; none of these peaks was present. This would indicate that the ®°Mo activity has to be at least a factor of 100 smaller than the °°Zr ac- tivity. The same holds true for all the tellurium isotopes; neither the relatively short-lived ***"Te nor the longer-lived '**"Te could be identi- fied. A further confirmation of the absence of tellurium was the fact that we could not identify **?I. Since '°*I has a 2.3-hr half-life, this would 132 T then indicate the absence of the longer half-lived e. LEspecially the absence of *°?1 is an important proof because this nuclide can be identi- fied more easily than any other nuclide in a gamma-ray spectrum. (Ve did 131 133 I I identify in the drain tank all longer half-life iodines such as and '%°1.) 3 ’ As in the case of molybdenum, the activity of antimony, tellurium, or 1327 had to be at least a factor of 100 smaller than the ®°Zr activity to go unidentified in these spectra. Table 7.6 shows the major nuclides identi- fied at various times after reactor shutdown. There were several peaks in 232 the spectra that could not be identified, probably due to the U chain decay products. 99 Table 7.6. Major nuclides identified in fuel salt in the drain after November shutdown ®8Kr (Rb) 13371 ler 1351 9szr 135Xe ?5Nb 137¢Cs ®77r t4%Ba(La) *7Nb 41Ce '93Ru (very small activity) 143¢Ce 1°Ru(Rh) (very small activity) 14%Ce(Pr) 1311 7.6 Group G Spectra These spectra were recorded during the November shutdown period on the heat exchanger, the main off-gas line, and the fuel lines. The portable maintenance shield was used for the survey. The results presented in this section are those we were really looking for. The data recorded in July were useful for the analysis of the long- half-life nuclides, and, above all, it served as a very profitable shakedown of the equipment. It should be borne in mind that all results are extrapo- lated to reactor shutdown time by simple exponential techniques. The rea- son for this extrapolation was to compare results from spectra taken at different times. This extrapolation might yield, however, a maximum ac- tivity value for some isotopes that is too high because of the decay of a precursor nuclide. For example, the extrapolated value of *>'I might be too high by a few percent because of the decay of *®Te. We will discuss this in detail in Chapter 8. Although the analyzer broke down about 24 hr before this planned re- actor shutdown and drain, everything was made operational in time. The 100 spectra recorded through the shield plug holes in the first two days after the reactor shutdown while the portable maintenance shield (PMS) was being set up are reported in Section 7.5. The reactor shutdown and drain occurred on November 2, 1969, at 1441 hr. Never before was the PMS installed so soon after a reactor drain, However, this did pose some problems: the radiation coming from the open reactor cell was considerable, and extra precautions had to be taken in installing the PMS at the location where the two lower shield plugs had been removed. However, no one received a radiation dose above the per- mitted level. During every day shift, the PMS was installed over a dif- ferent part of the heat exchanger (or from time to time over the jumper line of the main off-gas line); spectra were then taken during the evening and night shifts. As reported earlier, we could scan with one PMS setup approximately 35 in. along the longitudinal axis of the heat exchanger or the whole length of off-gas jumper line. Spectra were taken at intervals of 1 to 2 in. along the heat exchanger axis and 1 in. along the length of the jumper line. Based on our previous experience, we judged it worthwhile to increase the real counting time to 800 sec per spectrum. This proved to be advantageous for the counting statistics and allowed us to identify smaller peaks better. Once a 35-in. scan was finished, we set the detector over a specific spot and recorded several long counts of 10,000 sec real counting time for an even better analysis of the nuclides present. These long counts were continued until the day shift was ready to move the PMS to the next position. The entire survey along the length of the heat exchanger was done twice. Several areas were scanned a third time; the off-gas jumper line was scanned three times. No transverse scans were made over the heat ex- changer. It took approximately three weeks to finish the complete survey. It is worthwhile to note that spectra taken of the same spot but at dif- ferent times after reactor shutdown did, in general, yield values that were very similar; this gives some confidence in the results obtained. There were two series of spectra that gave higher nuclide activity results. These were taken with the 1/16-in. hole collimator insert at the very beginning of the survey. We estimate that these slightly higher re- sults were due to two problems: 101 1. The 1/16~in. hole in the collimator was far from straight and proved to be difficult to calibrate in relation to the 1/8-in. hole col- limator; its calibration value, hence the detector system efficiency curve for this case, might have been somewhat in error. 2., The cacophony of nuclide photopeaks shortly after reactor shutdown was such that it was obvious that photopeaks from several shorter- and longer-half-life nuclides fell on top of each other, thus causing the com- puter analysis program to overestimate some nuclide activities. As much care as possible was taken to ensure that the activity calculation con- sidered only those photopeaks that had little or no interference from other peaks. This was, however, not always possible for the spectra taken in the early stages of the survey. 7.6.1 Heat exchanger About 241 spectra were taken along the longitudinal axis of the heat exchanger. With a few exceptions, all could be analyzed successfully. The spectra taken over the heater connector boxes were discarded; the extra amount of shielding material that was obviously present attenuated the gam- ma rays too much to allow intelligible conclusions from those spectra. The results are presented per nuclide and in disintegrations per minute per square centimeter of heat exchanger tube. Data in which we have less con- fidence because of the above reasons are indicated in the figures as open circles. Niobium~95 (Fig. 7.21). As was also found from the spectra taken in July 1969, there appears to be an ipcrease of activity near the baffle plates. There was a good agreement between the data taken in July 1969 and those in November 1969. All results were based on the single °°Nb photo- peak of 765.8 keV. The range of activity is 0.090 to 0.27 x 10'? dis min~' cm”? Molybdenum-99 (Fig. 7.22). Because of the relatively short half-life of °°Mo, a large portion had decayed before we could survey the whole heat exchanger. There is, however, ample reason to believe that molybdenum would deposit all over the heat exchanger. dis/min +cm2 OF HEAT EXCHANGER TUBE ORNL-DWG 70-7366R T 1 ! ] T | | | | e e 0.30 Ef2 | Po | ! 1 P ° ‘... O.L 0.20 E12 - .. o . e { . o e .. ! - ...' . ..01-.‘ ° * I ‘o...O oo 0" ..n‘* .. % o : L 010 Et2 \ B oo t ¢ { l ode o . P "‘3 s s o . L . | | . promm e} s N ; L_ | | ‘ 1 : | HTR. PLUG HTR. PLUG ‘ SPACER HTR. PLUG | SPACE!R ‘ R | - ‘ - POSITION HOLE IN ‘ SHIELD PLUGS Fig. 7.21. Activity of 95Nb at reactor shutdown on November 2, 1969, in the MSRE heat exchanger. ¢0T ORNL -DWG 70-9574 < . ! T | 1 r w w 0.50 B2 ¢ i | e | i [ T M | . = | 1 I ol ° o ' &' 0.40 EfR i | x P Lo o | - 1 | 0 # . 0 .# . | . a® NE & ; ; o c f ' ) 030.30512' | | ¢ .| el e e = | . 1 i g % 0.20 Ef2 - i : S i . i ’ N E % . ' t o - . » » l i fEomee | S Tt | S | ‘ ] i o ; | ‘fi | | ?fi | HTR. PLUG | SPACER HTR. PLUG t ER | 1 . HTR. PLUG | FUEL SALT FLOW — | o POSITION HOLE IN I SHIELD PLUGS Fig. 7.22. Activity of ?’Mo at reactor shutdown on November 2, 1969, in the MSRE heat exchanger. €01 104 Although there is considerable scatter of the results (due to poor counting statistics). it does appear that there is again a concentration of activity near the baffle plates. Results were primarily based on the average activity yielded by the 739.7- and 778.2-keV photopeaks. The rather steep drop in the detector system efficiency below 250 keV made the 140.5-keV photopeak less desirable for analysis (see Chap. 4). The range of activity is 0.15 to 0.40 x 10?2 dis min~* cm~?. Ruthenium-103 (Fig. 7.23). Essentially the same distribution of the ruthenium deposition occurs as with the previous nuclides. The increase in activity near the heat exchanger tube sheet or near the baffle plates is somewhat more pronounced than with °°Nb or °°*Mo. The results were primarily based on the 496.9-keV peak and to a small extent on the 610.2-keV peak (for confirmation only). The range of ac- 5 ~ - tivity is 0.10 to 0.70 x 10 dis min™* em™* RutheniumRhodium-106 (Fig. 7.24). 1In addition to the increase of activity near the baffle plates, there is a good deal of scatter in the data; this might be expected because of the longer half-life of '°°Ru and hence its smaller disintegration rate. For the same reason, one should not expect this nuclide to be even close to the saturated deposition concen- tration. In general, the resultis were based on the average of the 511.8- and 621.8-keV photopeaks. The range of activity is 0.050 to 0.14 x 10*! dis min~* cm™? Antimony-125 (Fig. 7.25) Antimony-125 has a long half=life, a small fission yield, and photopeaks that could be easily confused with those of other nuclides, all circumstances which are not particularly advantageous for a good gamma-spectrometric analysis. d 125 Contrary to the molybdenum results, we detecte Sb mostly near the end of the survey when most of the shorter-half-life nuclides had disappeared. i258 Sb was Figure 7.25 presents the results for those locations where identified; it is, however, believed that this nuclide was distributed in comparable amounts all over the heat exchanger. The scatter in the data is largely due to the bad counting statistics., dis /min - cm2 OF HEAT ORNL-DWG 70-7364A e * | [ | I T J 2 - i | e $ -of e _ 060 EM e b . pet —~ - { — | | o . ..:. — . i . ?' - . ."f"_l % O 40 E“ Coeee J. " + - - """ " - — T — o ."I”**'Q—"’*—*"""” T —.—.—.T' li g T s 02N gl ey m % - 020 E{f t--om - @ o - ,"gi"..nn | e _ .. | : L P e . L>f] . - .eg. | . ] 00 . : ..':$ e I O ... TI B I | ! | | | | | | i | | . | HTR. PLUG HTR. PLUG l SPACER HTR. PLUG E SPACER | ! | | H| FUEL SALT FLOW —» | POSITION HOLE IN ‘ SHIELD PLUGS Fig. T7.23. Activity of 1038y at reactor shutdown on November 2, 1969, in the MSRE heat exchanger. 0T ORNL- DWG 70- 95?5 '<_I ) 1 T T ST T 1 W& 0.20 EM ! J | | e 1 | | > | | o | | ‘ y u ; | | o I | © g 045 EH | | ] L -4 E . N B \ | Ao o el T ] °©Z 010 Eft| e | |, ces e # S ,| £ 5 i et Tl e T e, et e e ] e <= %5 0.05 Eff *e, | “ | . ll eee e - I o |r . o Lo e ] I 1 S | o ° | E | I | | | | | | | HTR. PLUG HTR. PLUG } SPACER HTR. PLUG ; SPACER I | | — . = ] - FUEL SALT FLOW —~ | - - POSITION HOLE 1NT SHIELD PLUGS Fig. 7.24. Activity of '9Ru-Rh at reactor shutdown on November 2, 1969, in the MSRE heat exchanger. 901 E"[ ORNL-DWG 70-9576 WL T ( ! | l . | T ® 030 E10 - | } | . i b I-:? . i ; . ‘ i o o | ‘ .‘j . o S 0.20 E10 - | j | . . -, . . 7= | . [ | s . * £ < 040 E10 - | ; | . E < * | | et | W | | | . 5 O | 1 - — 1 | ! | | | | f HTR. PLUG HTR. PLUG ; SPACER HTR. PLUG , SPACER g | R L 0 Fig. T7.25. i FUEL SALT FLOW —= POSITION HOLEIFfiJ SHIELD PLUGS 125 Activity of ob at reactor shutdown on November 2, 1969, in the MSRE heat exchanger. L0T 108 Results were based mostly on the 427.9-keV photopeak and, to a lesser degree, on the 600.8-keV peak. The range of activity is 0.080 to 0.30 x 10*° dis min~' cm™?. Antimony-126. This nuclide proved to be difficult to identify. Its fission yield is relatively small, and its two major photopeaks (666.2 and 695.1 keV) fall together with those of '°?I and ***"'Te. The next important peaks were also close to photopeaks of other nuclides. Although we identified '*°Sb several times, the scatter in the com- puted activity is such that the accuracy is expected to be low. A nuclide distribution pattern along the heat exchanger could not be established. The range of activity is 0.023 to 0.14 x 10*° dis min~' cm™?; average ac- ?; and standard deviation of the average tivity is 0.89 x 10° dis min~' cm™ activity is 0.11 x 10°, Antimony~127 (Fig. 7.26). The increase in activity near the baffle plates, although apparent, is by far not as strong as with °5Nb or *°°Ru. Although '?7Sb could be detected in almost every spectrum, we had to rely almost entirely on one photopeak (684.9 keV) for the activity analysis; naturally this causes some scatter in the data. The range of activity is 0.20 to 0.55 x 10'* dis min™" cm™?, Tellurium-129n (Fig. 7.27). The amount of *?*Te deposited on the metal surfaces might be of crucial importance in explaining the deposition mechanism of several of the decay chains. There are, however, several problems that put a strain on an accurate analysis of the *?*°Te (see Chap. 4.3). The scatter of the results is considerable, although most within a certain range. There does not appear to be a strong increase of activity near the baffle plates; one might observe, however, a slight increase of activity near the tube sheet of the heat exchanger., The range of activity is 0.10 to 0.30 x 10*' dis min™' ecm™*%. Iodine-131 (Fig. 7.28). The increase of activity near the baffle plates is definitely there, although much less pronounced than with the 103 R metals °*Nb and u. The photopeaks were easily identifiable, and the confidence in the data is good. For the analyses we used the 364.5- and dis/min-cm? OF HEAT S 3 IR I ORNL-DWG 70-9577 L FUEL SALT FLOW —» o e T - I T L 060 £ | o e i e ~ . | ., ot B e e e T 040 Ef e 1 Tee Lt Rt .ou e et e 2% L ‘ " oo s ws ¢ T % o . | s W, . g Lo - . o ' « e | - e *e I . —e- — o g "o n — . 2 020 E o’ | L% : - | T Y . ! | o ; o P 5 w . | b —— _ : > ' ! f ' : W 0 — 1 - i e i i | i i : | { | | ! _ HTR. PLUG HTR. PLUG } SPACER HTR. PLUG SPAC%F\’ i e i - : ! ) T | ; . - b POSITION HOLE IN’I SHIELD PLUGS Fig. 7.26. Activity of '27Sb at reactor shutdown on November 2, 1969, in the MSRE heat exchanger. 60T 2 W 040 EY4 e e : s e . | Ll m ol | I o 1 o | . v o R —te w " 030 EN ' L P . ! . Sk a | . .o - o L i . ! . ® T LT . Twgl e T, . A NE (ZD 020 EN { * ’ * l) . -~ :f See 0 . '*.{.'.'.‘Tl' - . & 77; e (‘-) on c o i 2 5 @ o @ £~ S o 5 ™~ < £ 00 © © = ut a 5 J [ o = 108 ORNL-DWG 70-9605 N 0.2 Fig. 7.39. T T — o !"‘* - ‘331{0 529.9 kev ‘ 1 o T - - - So 847 0key T T ‘ ; | 134104 884.1 kev | | 1l | $1072.5 kev | ; ‘ I 5 - + { . . t | . o o e 1 | q | | | © o i O‘ + 0 } - e T: - — 5 | O C I 'l T o 1 [ . T Wfihfif *’—_H e} ‘ -2 ? I . e v e - e E - - 0 ; ; o 0 ! . oo s 1 51 1 b L - o 9 ° | * o 4 Tm——f ——— % —— e —— i i . i _ _. _ o _1 \ ! i i ‘ ! ‘ | | 1 ] ; : ‘ | 1 0.4 06 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 DELAY TIME AFTER REACTOR SHUTDOWN (days) Activity of '337 and '3%I after final reactor shutdown in the MSRE primary heat exchanger (through the shield plug hole). et 134 Figure 7.40. The activity of '2%%Xe was again appreciable, with a maximum of 0.60 x 10'! dis min~* cm™2. It might be of interest to note that during the very first spectra taken after reactor shutdown, there was still some salt in the heat ex- changer (in the course of draining away). These spectra contained several 1337 Once the salt photopeaks that could be identified as coming from was drained from the heat exchanger, those photopeaks disappeared entirely, indicating that at least '®°I1 and, in all probability the other iodines, remain with the salt, It should also be noted that °'Sr was also identified in many of these spectra, 7.8 Group I Spectra These spectra were taken after the final reactor shutdown from samples of the reactor cell air after there had been indications of an increase of the cell-air activity.- Results are reported separately.®® 7.9 Group J Spectra Spectra were taken from the coolant-salt radiator a few days after the final shutdown and drain of the fuel- and coolant.salt systems. The ob= jective was to determine if any radioactive corrosion products in the coolant salt were depositing in the coolant radiator. Since some of these corrosion products might have been activated by delayed neutrons in the primary heat exchanger, those corrosion products in the coolant-salt radiator could possibly be observed with the gamma-ray spectrometer. The detector was set up in front of the radiator with the radiator doors openj no collimator was used. Since the MSRE main off-gas 18p. H. Guymon et al., Preliminary Evaluation of the Leak in the MSRE Primary System which Oceurred During the Final Shutdown, internal memo- randum (April 197Q). S NUCLIDE [(dis/min}/cm2 of heat exchanger tube] S, 5 - U N wn ~ ORNL—DWG 70— 9606 1 _ I ] ] B 1 T . T I o : ‘ - f C— + e e o e - o, SOV IR | | ; "39xe:0 249.5 kev ! o | L e ....’ -y . I— o ; o © 1 | ‘ o ! ! 1 ’ o - o | : O 0.2 DECAY TIME AFTER REACTOR SHUTDOWN (days) Fig. 7.40. Activity of '?°Xe after final reactor shutdown in the MSRE primary heat exchanger (through the shield plug hole). o | A m_’___mfi*fi___mm | ; ! ‘ | ] i | . : | ‘ . i i { ! l \ . o . _ L ‘ ] | 0.4 06 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 CET 136 line, although heavily shielded, passes not far away from this detector lo- cation, it appeared that the majority of the recorded photopeaks came from the off-gas line. Typical examples of identified nuclides that should not be expected in the coolant radiator were: °°Nb, °°Mo, *°°Ru, *°°®Ru(Rh), 1317 132pa_1, '3%%Xe, '*7Cs, and '“°Ba~La. These nuclides formed the major peaks in the recorded spectra. Some other nuclides could be identified with some degree of confidence. An approximate relative activity has been calculated for these in relation to the activity of °°Co (Table 7 13). Because the unshielded detector obviously picked up gamma rays from different locations, the identified corrosion product radiations do not necessarily need to have come from the coolant radiator. A collimator would have been useful in identifying the sources of the activities; however, the activity level was too low to permit use of the available collimation equipment. 7.10 Group K Spectra The purpose of this study was to measure the flow rate of coolant salt in the loop. Certain components of the coolant salt will be slightly acti- vated by delayed neutrons in the primary heat exchanger to produce *®N and '?F. By measuring their activity at two places along the coolant-salt line, it is possible in principle to determine the flow rate from these decaying nuclides. The results of these measurements have been reported elsewhere.'’® 7.11 Group L Spectra A graphite sample was lowered into the fuel pump bowl through the sampler enricher; this sample was left for several hours in the fuel pump bowl and then moved back into the sampler enricher. A special arrangement *°C. H. Gabbard, Reactor Power Measurements and Heat Transfer Performance in the MSRE, ORNL-TM-3002 (May 1970). 137 Table 7.13. Relative activities of corrosion product nuclides found in the survey of the coolant radiator These nuclides need not necessarily be in the radiator Confidence in Disintegration rate Nuclide nuclide identification relative to °°Co ®°Co Good 1 2%Na Reasonable 0.06 *icr Low %2.5 *°Fe Good 0.4 S5Ni Low z3,5 138 of a collimator with the detector was required to record gamma-ray spectra from this sample as soon as it was back in the sampler enricher. The re- actor was at full power during this experiment. The objective was to study the deposition of short-lived fission products on graphite. A similar ex- periment was performed with a sample capsule filled with fuel salt as well as with a dummy sample capsule without salt., These experiments have yet to be analyzed in detail. 7.12 Group M Spectra Several miscellaneous spectra were taken from the lube o0il system, the 523 off-gas line, and the roughing filters. Argon instead of helium was used as a reactor system purge gas for some special experiments to study the influence of a different gas on the stripping efficiency of noble fission gases in the fuel pump bowl as well as the change in the bubble fraction in the fuel salt. Since part of this argon could circulate with the fuel salt in the reactor system, some argon activation was to be expected. As a result, we even found some “*Ar in the fuel pump lube o0il system during this experiment. This was proved by placing the detector near this lube o0il system. Apparently “*Ar was entrained by the lube o0il from the fuel pump to the lube o0il system outside the reactor cell, Shortly before the reactor shutdown on June 1, 1969, a plug developed in the main reactor off-gas line. Purge gases were then exhausted through the pump bowl overflow tank and its off-gas line (line 523). A plug also developed in this line after a period of time. Through other methods it was established that the plug was in a particular section of the 523 line. In an effort to locate the plug precisely, the gamma-ray spectrometer was used to survey that pipe section. This method would have been viable if the plug had consisted of radiocactive material. Although there was some increase of activity near a valve in the line, we could not identify the real location of the plug. Afterward, when this pipe was taken out and inspected, it was learned that the plug consisted of carbonaceous material; hence it is not surprising that we did not detect any activity increase, 139 The nuclides identified in this pipe section were similar to those found in the main off-gas line such as noble metals and decay products from noble fission gases. The roughing filters, or prefilters, are used for the containment ven- tilation system. The ventilation areas normally comprise the reactor high- bay area but also include the reactor and drain tank cells during mainte- nance operations when the reactor cell containment is opened. During peri- ods of maintenance, one can expect some activity from the reactor cell (main- ly activation products) to collect on these filters. During replacement of the roughing filters, some gamma-ray spectra were taken to check on the nature of the retained activity. Table 7.14 shows the nuclides identified from these spectra together with their activity in relation to °°Co. The spectra were recorded many weeks after use of the filters, so no shott- lived activities should be expected. 140 Table 7.14, ©Nuclides identified from gamma-ray spectra recorded from the containment ventilation roughing filters Confidence in Disintegration rate Nuclide identification relative to ®°Co ®°Co Very good 1 5%Fe Very good 0.54 Sicr Very good 0.50 *“Mn Questionable 0.55 120Mp g Very good 0.17 193Ru Very good 1.32 198Ru-Rh Very good 1.89 ?3Nb Very good 0.77 '37Cs Reasonable 0.18 1827y Reasonable 0.085 2%Na Reasonable 0.005 12%gh Questionable 0.069 141 8. CONCLUSIONS The primary purpose of this study was to compile, in a readily usable form, a block of data on certain aspects of the behavior of fission products in the MSRE. We believe that the results given in the previous chapters are a fair representation of the analysis of all the recorded spectra that per- tain to the fission product deposition in the parts of the MSRE that could be examined by this method. In general, we have found remote gamma-ray spectrometry to be readily applicable and productive of much useful information about the MSRE. Many observations (e.g., possible discovery of new gamma rays, fundamental evalu- ation of collimators, etc.) are believed to be in disciplines that are not directly connected to the fission product behavior of this reactor and con- sequently are omitted from this report,. We realize that this work is only part of a very complicated matter in which the coordination of several specializations might be required to develop a more complete picture of fission product behavior in the MSRE, Hopefully, the information presented in this report can be combined with other data to achieve that goal. This chapter presents some of our observations about the performance of this study. In addition, we draw some conclusions concerning the be- havior of certain fission products. 8.1 Data Collection and Analvysis 8.1.1 Gamma spectrometer system One of the main reasons that we were able to analyze these complicated spectra successfully was the excellent resolution capability of the de- tector. Since we had to collimate and shield the incoming gamma rays con- siderably, there was no need for a large, expensive, high-efficiency de- tector. For these studies a medium-sized (26-cm®) and hence rather inex- pensive high-resolution crystal sufficed. The need for a multichannel * It should be noted that although the detector was only of medium size, it had a rather high peak-to-Compton ratio (27:1), thus permitting detection of many low energy gammas that would not otherwise have been observed. 142 analyzer (4096 channels) with magnetic tape recorder seems indispensable with such a detector. Since our spectra were complicated, the spectrum analysis could only be done efficiently by computer. The computer program was complex and required a large memory, 610 k-bytes. It is doubtful that a smaller computer coupled directly to the analyzer system would have served Our purpose. The collimator assembly served its purposes well; with the proper choice of a collimator insert, we could limit the total amount of radiation to the detector without excessive use of shielding material. Also, we were able to observe fission product depositions in reactor components at very definite locations. Since the assembly was mounted with three adjust- able set screws on an axial thrust bearing, instead of directly on the portable maintenance shield, we could easily make small adjustments to the assembly orientation. However, it was difficult to make a straight col- limator beam hole, The idea of a free-~floating thick-walled stainless steel precision tube around which the shielding lead is poured can be fruitful if some more thought is given to it. The locational equipment, that is, laser and surveyor's transits, per- formed very satisfactorily and proved to be precise, reliable, and inex- pensive. Since we moved the detector frequently, it is felt that any so- phistication of this equipment (e.g., such as mirrors to guide the laser beam around the detector) would have been detrimental to its reliability. 8.1.2 Calibration Very little is known about the efficiency of gamma-ray detectors in relation to a collimated beam. This, together with the complexity of the heat exchanger geometry, led us to believe that the effort given to the empirical calibration of equipment was worthwhile. The fact that differ- ent spectra recorded from the same spot on a component, but taken with en=- tirely different shielding configurations, yielded virtually the same ac- tivities lends confidence to the calibration results obtained. Silver-110m proved to be a good calibration source. If one has re- actor irradiation facilities available, such a source can be made rather easily. The extension of the efficiency curves, as obtained with **°"Ag, to both the lower and higher energy ranges was possible by using the actual 143 fission product spectra obtained during the experiment itself. The rela- tive efficiency curves obtained from the fission product spectra were consistent with the absolute efficiency curves found with the calibration source. 8.1.3 Computer analysis program Considerable time and effort were required to convert the computer analysis program, which originated at Lawrence Radiation Laboratory, for operation at the ORNL computing center. The program is quite powerful and, generally speaking, satisfied our needs well. For application to spectra like ours, where there are so many multiple photopeaks, however, we can en- vision several improvements, particularly in the analysis of multiplets. Our table of radionuclides, as used in the library of the computer program, was a very useful todl for the analysis of the spectra. This table might be also of interest to others working in the gamma-spectrometry field, and so is presented as Appendix A. Although this table contains the latest published data, we realize that it is far from complete. 8.2 Results — General We will draw some general conclusions concerning the fission product distribution in several reactor components, without, however, going here into detail on the different decay chains. 8.2.1 Metal surfaces in direct contact with the fuel salt Metallic fission products, such as isotopes of niobium, molybdenum, ruthenium, rhodium, antimony, and tellurium, were present on the walls of the reactor system that were in contact with the fuel salt. Their depo- sition on the walls represented, in general, a large fraction of the total amount of that nuclide present in the entire reactor system. Because of additional metal surface exposed to the salt near the baf- fle plates in the primary heat exchanger, one might expect a small increase in activity there. It appears, however, that the observed large increases in activity near these baffle plates (two to four times higher than in intervening areas) are due to additional effects. It may be that the salt flow pattern influences the deposition rate of these fission products. For example, niobium, molybdenum, and ruthenium-rhodium exhibit a higher 144 activity increase near the baffle plates (2.5 to 4 times) than antimony, tellurium, and iodine (1.5 to 2 times). A study of the activity of the several iodine isotopes directly after reactor shutdown and subsequent fuel drain during the final reactor shut- 131 down in December revealed that the iodine activities ( I and especially '327) build up after reactor shutdown. Although we were never able to detect **°I after the drain of the fuel, we did detect it during the drain. Therefore we conclude that the observed large iodine activities in the empty system were mostly due to the decay of their precursors that deposited on the wall; iodine itself, however, remains with the fuel salt. In comparing the relative activities of the deposited metal fission products (relative to °°Nb) in the heat exchanger and in the fuel lines (101 and 102), there is no major difference in the deposition rate on the heat exchanger (between the baffle plates) and on the fuel lines. Molybdenum and ruthenium are possible exceptions; their relative activity appears to be somewhat higher in the fuel lines. It should be stated, however, that this is a tentative conclusion since we did no calibration work on the entirely different geometry of the fuel lines. The amount of activity in the heat exchanger caused by the decay of noble fission gases is appreciable. Although most gas activities disap- peared after a few hours because of decay and some circulation of purge gas, their decay represents a significant heat source during the first hours after drain of the fuel salt. These noble fission gases probably came from several sources: 1. A release of gases absorbed in the graphite bars in the reactor core, 2. A release of gas bubbles from the fuel salt while in the course of draining to the drain tank. 3. A back-flow of gases from the holdup gas volume of the main reactor off-gas line. In order to drain the fuel salt, a gas pressure equalizer line is necessary between the drain tanks and the fuel sys~ tem; this line (521) ties into the main reactor off-gas line at the 4-in,-ID gas holdup section. A calculation of the volume of these back-flowing fission gases during drain largely explains the unex- pectedly high activity of noble fission gases in the reactor system after shutdown. 145 It is reassuring to note that the magnitudes of the activities de- tected in the heat exchanger during the July shutdown were very close to those detected during the November survey. 8.2.2 Main reactor off-gas line As might be expected, the major activities in the main reactor off-gas line during reactor operation are due to the decay of noble fission gases and their decay products. These nuclides still form a large source of ac- tivity shortly after reactor shutdown, but they decay rapidly and are di- luted by purging of the empty fuel system. Metals such as niobium, molybdenum, ruthenium, rhodium, antimony, and tellurium (iodine) could barely be identified during reactor operation but formed the major activity source soon after reactor shutdown. Since we were not able to detect '?°I in the off-gas line, although other iodine isotopes with metallic precursors were identified,* we believe that iodine proper does not separate from the salt and will neither disap- pear in large quantities into the off-gas line nor remain with the metal walls when the system is filled with circulating salt. We were not able to positively identify any nuclide, such as °°Zr, that is supposed to remain with salt; that is, we could not detect any ap- preciable quantity of fuel salt in the off-gas line. Although on a few occasions the computer program found a photopeak possibly due to *°Zr, we scrutinized the actual plots of these spectra and were not convinced of the presence of salt within the range of our nuclide detection sensitivity. Concerning the off-gas line geometry, this meant that we could not detect activities less than 0.5 x 10'*° dis min™" in.”! or an activity less than 5 x 10=* of the detected °°Nb activity. Taking into account the purge-gas flow rate through the jumper line, it is not surprising that we did not detect a decrease in nuclide activity along the jumper line. % The precursors of **°I have half-lives too short for significant amounts to escape from the salt. 146 Comparing the activities calculated from the July and November surveys, the longer-lived nuclides, such as '°°Ru-Rh and '*’Cs, seem to be con- sistent; °°Nb, ‘°?Ru, and '?°"'Te appear to be about four times higher during the November survey. This could partly be explained because of the power history before shutdown as well as by the fact that the main off-gas line was partially plugged during the latter part of the power operation before July. (At that time the off-gases were routed through the off-gas line of the fuel pump overflow tank.) Shorter-lived nuclide activities (e.g., **'I) would be even more influenced by this plugging problem. 8.3 Elements and Nuclide Chains Although it is clear that the reported results are only a part of the total picture of fission product behavior, it is possible to draw some ten- tative conclusions concerning the behavior of certain elements and nuclide chains. 1In particular, it is of interest to examine the behavior of iso- topes that belong to the same element or to the same decay chain. 8.3.1 Niobium Both ®°Nb and °’Nb were identified in the heat exchanger after the November and the final shutdown. Because of the short half-life of °’Nb as well as its precursor, the disintegration rate of this nuclide reflects only the deposition of niobium shortly before shutdown. The °°Nb activity would represent a much longer history where chemical conditions of the salt were different. Because absolute activities derived afterxr the final shutdown may be uncertain, we have compared the ?’Nb results for the November and final shutdowns by normalizing them to the °°Nb values. Table 8.1 shows these results corrected for decay to reactor shutdown time,* Also shown are half-lives and fission yields. These data pertain to the area in the heat exchanger under the shield plug hole. * Since it is niobium that deposits on the metal walls or disappears into the off-gas line, it appears legitimate to extrapolate back to reactor shutdown time to obtain the maximum niobium activities. 147 Table 8.1. Comparative values of °°Nb and °7Nb Activity relative to °5Nb activity in heat exchanger Fission yielda November shutdown Final shutdown (%) Half-life ?SNb 1 1 6.05 35.5 days 7Nb 0.63 3.0 5.62 72.0 min 91t is estimated that the total fission rate in the MSRE was due to the contribution of the following isotopes: 233U, 94%; 2°°U, 2.25%; 23°pu, 3.75%. Actual fission yields were calculated from B. M, Rider, A Survey and Evaluation of Thermal Fission Yields, GEAP-5356. 148 Concerning the activity results after the November shutdown, one might expect the °’Nb activity to be in equilibrium with its environment whereas the ?°Nb activity is not; the uninterrupted power run was less than two months. A possible explanation for the November shutdown ®’Nb/®°Nb activity ratio might be to assume a time lag between formation and deposition. An average time lag, or residence time in the salt before deposition, of ap- proximately 100 to 150 min would explain the calculated ratio of 0.63. The small ?°Nb buildup during the short power run after the long shut- down period in November, as well as two beryllium additions prior to the final shutdown (which are known to affect the niobium behavior), might ex- plain the rather different activity ratio of 3.0 after the last shutdown. Currently it is thought that the oxidation potential of the salt, which is influenced by the beryllium additions, affects the niobium solubility in the fuel salt, This, of course, forms another dimension to the deposition problem. Therefore, a study of the °’Nb concentration in the salt in re- lation to the oxidation potential might be quite useful in future reactor systems. The reason for using the °’Nb concentration is that it is not in~ fluenced by a long reactor power history as is °°Nb. A simple gamma-spec- trometry assay done at the reactor site would be ample. We were not able to positively identify ’Nb in the main off~gas line; there were indications of this nuclide, but too few to assign a numerical value to it. 8.3.2 Ruthenium-rhodium Three ruthenium nuclides were identified in both the heat exchanger and the off-gas line: *°?Ru, *°°Ru, '°°Ru-Rh. Since it was assumed that the rutheniums are the first nuclides in their respective fission decay chains that deposit on the metal walls or escape into the off-gas line, an extrapolation to reactor shutdown time would then be legitimate to ob=- tain their maximum activities, This might not be the case for *°®Ru, since its precursors, *°°Mo and ‘°°Tc, also will deposit on the metal walls; it would, however, not introduce a large error in our evaluation of the maxi~- 105 mum Ru activity. Table 8.2 shows comparative values for these three isotopes. The activities are given relative to the activity of *°°Ru-Rh. 149 Table 8.2. Comparative values of three ruthenium isotopes identified in the reactor system Activity relative to '°°Ru-Rh Heat exchanger, Off-gas line, Heat exchanger, Nov. shutdown Nov. shutdown Final shutdown Half-life Fission yielda 103p. 3.6 7.4 1.2 40 days 1.80 1°5Ru 6.0 1.3 6.2 4.43 hr 0.60 "°®Ru-Rh 1.0 1.0 1.0 367 days 0.41 aIt is estimated that the total fission rate in the MSRE was due to the contri- bution of the following isotopes: 223U, 94%; ?3°U, 2.25%; *°°Pu, 3.75%. Actual fission yields were calculated from B. M. Rider, 4 Survey and Evaluation of Thermal Fission Yields, GEAP-5356. 150 Let us look first at the *°?Ru and *°®Ru-Rh. Taking into account the 103 h whole power history of the reactor one can calculate how muc Ru and *°®Ru are present in the entire reactor system: in the salt, on the graph- ite the off-gas line, or on the metal walls. These calculated inventories 103 106 yield a ratio of the Ru activity relative to the Ru~Rh activity of roughly 6.5 both for the November shutdown and the final shutdown. Because of the difference in decay half-lives of these isotopes, this ratio should decrease with increasing age of the mixture. Tentatively one would conclude 103 106 R_ from the November shutdown ratio of 3.6 that the u Ru~Rh mixture has an average age of several weeks or that ruthenium does not tend to deposit very readily on metal walls. The same ratio calculated for the off-gas line is 7.4. This seems to suggest that ruthenium separates from the salt and disappears into the off-gas line (or possibly deposits on the graphite rather than on the metal walls). 1In other words, since we could barely detect ruthenium in the fuel salt in the drain tank, it appears that this element has only a slight ten- dency to deposit on metal surfaces and rather will disappear into the off- gas line (or deposit on the graphite) Data on the deposition on graphite would be very important for a complete picture of the ruthenium behavior. 105 103 If one assumes that Ru deposits by the same mechanism as Ru and '°®Ru, the '°°Ru data from Table 8.2 are rather hard to explain. One al- ternative possibility would be to assume that it is the short-lived *9%Mo 105 105 or Te that really deposits, the observed Ru in the heat exchanger being mainly the result of its precursors deposition. A check on the ac- tivity of ‘°’Ru relative to “°Mo both in the off~gas line and the heat ex- changer does not exclude this possibility. Because many other photopeaks are adjacent to or even coincide with its peaks, *°°Rh is difficult to identify. From a study of Figs. 7.8, 7.15 and 7.34, two conclusions may be drawn. The results concerning the heat exchanger, Figs. 7.8 and 7.34, suggest that the *°°Ru activity is 1053 equal to or higher than the Rh activity; the opposite is true for the off-gas line. Also, the '‘°°’Rh does not gradually build up from nothing 105 to a maximum in the heat exchanger. Apparently, Ru remains on the metal walls after its formation from *°°Mo-Tc. The *°°Rh seems, to certain degree 151 at least, to dissolve from the heat exchanger and then partly transfer to the off-gas line. In view of the *°°Ru—*°®Ru-Rh behavior we envision the following pos- sibility for the 105 decay chain: molybdenum deposits more readily on the heat exchanger than ruthenium; in agreement with the °°Mo data, it also escapes into the reactor off-gas line. The ruthenium does not desorb from the metal walls; therefore the observed deposits of this element on the heat exchanger and in the off-gas line are merely the decay products of the deposited molybdenum, Rhodium, formed by the decay of ruthenium, not only decays on the metal walls but also goes back into the salt and eventu- ally disappears appreciably into the off-gas line. The fact that only very small amounts of various rutheniums were found in the salt confirms the contention that these nuclides transfer rather quickly from the salt into the off-gas line (or onto the graphite). 8.3.3 Antimony—tellurium—iodine The identification of *2°Sb, *2?°Sb, *?7Sb, and **°Sb in the heat ex- changer and in the off-gas line leads one to believe that antimony is at least partly instrumental in the presence of tellurium and, as a result, also of iodine after drain of the reactor system. The emerging question then is: What governs the tellurium deposition? There are several possi- bilities, the extremes of which are (1) all tellurium, after formation from decay of deposited antimony, returns to the salt and moves elsewhere; that is, when the system contains fuel salt, no tellurium is present on the metal walls; (2) tellurium deposits or remains on the wall, Let us take an example of the fission decay scheme of the elements of mass number 131: 827% 1 131y 1317 s 152 Antimony-131 decays for 6.8% to the isomeric state of tellurium, *®'Te, and for 93.2% directly to the ground state, '*'Te; '*'Te decays for 82% 1211 and for the rest first to the ground state and then to directly to 1317 Table 8.3 shows the cumulative fission yields as well as the decay half-lives. In the first case assumed above, tellurium would remain on the metal walls only after the drain of the system., Because of the short half-life, the equilibrium concentration of '®'Sb on the walls would be relatively small, After shutdown, the iodine activity would build up rather quickly since both *®'Sb and '®'Te have short half-lives. The buildup of *3*'I through '?'Te would be small and provide roughly 5% of the maximum iodine activity. Since we did not detect any tellurium in the salt, the bulk of this element would then have to be either on the graphite or in the off-gas line, 131 In the second case, Te and '*'"Te would be also in equilibrium on the metal walls when the salt is still in the system; this means that their decay rates are governed by the respective fission yields. Since 131Mre has a much longer half-life than '®'Te, the equilibrium concentration of '31mTae is much larger than that of *>'Te. The buildup of '°'I is expected to come from three sources: “>'Sb, '?'Te, '?*YTe, 1In view of the fission yields and decay half-lives of these iodine precursors, roughly 70% of the maximum iodine activity would be due to **'"Te and the remainder to *°?Sb and '>'Te. The maximum iodine activity would occur approximately three days after reactor shutdown; however, this maximum would not be very pro- nounced, since the iodine activity from '®'Sb and *°'Te builds up quickly after reactor shutdown. Let us now examine the different figures and try to decide what actu- ally happened in the heat exchanger. Figure 7.18. Ve believe that the data based on the 364.5-keV photo- peak are the most trustworthy, although the values might be somewhat too high shortly after reactor shutdown because of a prominent °°Kr photopeak at 362.6 keV. Apart from the first data after shutdown, one would conclude from this figure that the *>'T does build up to a maximum activity two to three days after reactor shutdown. This observation would be in line with 153 Table 8.3. Information concerning the element-~131 fission decay chain Nuclide Half-life Cumulative fission yield® (%) t31gp 21 min 2.93 13177, 1.2 d 0.20 1317a 25 min 2.77 1317 8.07 d 2.93 Tt is estimated that the total fission rate in the MSRE was due to the contribution of the following isotopes: ?*°°U 94%% 223y, 2,25%; 2°®°Pu, 3.75%. Actual fission yields were calculated from B, M. Rider, 4 Survey and Bvaluation of Thermal Fisston Yields, GEAP-5356, 3 154 31T activity the above-mentioned second hypothesis. The relatively high in relation to, for example, °°Xb or °°Mo would also favor the second case. Figure 7,17 . If the first hypothesis were to hold, one would expect the **'"Te activity to be appreciably lower than the **'I activity because the amount of '°*'Te formed from *3'Sb would be quite small. This is defi- nitely not the case. Figures 7.18 and 7,19 . Antimony-132 has a decay half-life of approxi~ mately 2.1 min and *°'Sb approximately 25 min. (The decay half-lives and fission yields of these two isotopes do not appear to be very well known.) This means that their equilibrium concentrations on the metal walls, also taking into account the fission yields, differ by a factor of at least 8 to 10, Consequently, if the first hypothesis were correct, the ***I should be higher than the ***Te~I activity This is definitely not the case. Although it is fully acknowledged that a more precise analysis should be made of the antimony-tellurium-iodine data, it seems reasonable to con- clude from the above that both the antimony and tellurium isotopes do de~- posit on the metal walls while the fuel salt is in the system. Both of these elements seem to contribute to the total iodine activities. ITodine itself, however, will not remain on the metal walls as long as there is salt in contact with these walls. A study of the antimony, tellurium, and iodine activities of the other fission decay chains basically points to the same conclusion. 8 3.4 Extrapolation back to reactor shutdown time In order to plot and compare the recorded data during the actual sur- vey of the heat exchanger and main reactor off-gas line, all activities were extrapolated back to reactor shutdown time. This seems a reasonable procedure to obtain the maximum activities of the deposited elements. There is possibly one exception to this rule, *>'I. Because '’'I appears to at- tain its maximum activity onlv after roughly three days, the extrapolation would overestimate the maximum *°*I activity by about 20%. Thus, '°'I data given in Fig. 7 .28 and Tables 7.7 to 7.9 could be considered 20% too high. Since all data of the heat exchanger and main reactor off-gas line were taken during the actual survey, three or more days after reactor shutdown, 131 these extrapolated I activities are consistent among each other, 155 9. EPILOGUE We observed a rather large, unexpected activity due to the fission gases krypton and xenon in the heat exchanger. This could be explained by taking into account a flow of these gases from the main reactor off-gas line back into the reactor system during the drain of the fuel salt. Actu- ally, if one calculates the volume of the gases involved and their average age after fission, the observed activity in the reactor system can be ap- proximately accounted for. Where substantially larger activities are in- volved with large molten-salt reactors, a different layout of this gas equalizer line should be considered to avoid such an activity surge. With the exception of the iodines, all identified fission products migrated to the walls of the heat exchanger (and consequently to other metal walls in the reactor system), while the fuel salt was in the systemn. Unless the flush salt has entirely different chemical properties, it is difficult to expect that this salt would dissolve the deposited elements. With the exception of '*'I, *?3*I, and '°“I, the current flush salt is not expected to reduce appreciably the deposited activities; it still might be useful, however, to dissolve pockets of fuel salt or absorb radioactive dust. In the previous chapter we have tried to define a few causal relation- ships for the observed activities of deposited fission products. In view of larger molten-salt reactor systems, it might be useful — beyond academic interest — to study the recorded data in much more detail. TFor example, a detailed study of the niobium, molybdenum-~ruthenium-rhodium, and antimony- tellurium~iodine decay chains should be of interest since these elements seem to represent the major heat sources in the empty reactor system and main reactor off-gas system. After this satisfactory experiment, in which a sensitive gamma-ray spectrometer was used in a very intense radiation field, we feel confident that a similar technique could be very useful in larger molten-salt reactor systems. For example, a remote gamma-ray spectrometer aimed at a bypass line of the main fuel system could economically reveal very useful infor- mation: the ®’Nb/°°Nb ratio in connection with the redox potential of 233 the salt and the Pa concentration. 156 Although the gamma spectrometry system used in this study was in most respects highly satisfactory, two changes could be made that might signifi- cantly extend its usefulness. One change pertains to the detection of weak gamma lines (small photopeaks) in the presence of strong lines. Each gamma ray produces, in addition to a photopeak, a Compton continuum distribution with an energy range varying from O keV to just below the full energy of the gamma ray. A count in the Compton distribution arises when a photon loses only part of its energy in the detector. The remaining energy leaves the detector as a Compton scattered photon. Although the detection of low-energy gamma rays is favored because the counting efficiency rapidly increases as the gamma energy decreases, this effect is partly offset if higher-energy gammas are present. Because of the Compton distribution of the higher-energy gammas, the lower-energy photo- peaks will be masked to an extent that depends on the relative intensities of the various gammas. If they are of sufficiently low intensity the low- energy photopeaks will be completely masked. In the past few vears, a technique has evolved to greatly increase the peakfCompton ratio and thereby significantly extend the ability of the sys~- tem to detect low-energy gammas in the presence of high-energy gammas. This technique makes use of a Ge(Li) detector enclosed in a much larger de- tector called sn anticoincidence mantle. The mantle is often made of NaI(Tl), but can be a plastic scintillator. Ports are provided in the mantle to pro- vide access for the Ge(Li) detector and a beam of gamma rays. The gamma beam is allowed to hit the Ge(Li) but not the mantle. Gamma rays that lose all their energy in the Ge(Li) produce a pulse (count) in the full energy peak, Most of the gamma rays that scatter out of the Ge(Li) detector will be detected by the mantle. For such events, pulses are detected simultane- ously in both detectors. By proper electronic gating the spectrometer can be made to reject these coincident pulses and store only those that occur singly in the Ge(Li) detector. The resulting anticoincident spectrum has a very large full energy peak but very low Compton distribution. Because of the lower Compton distribution, the ability of the system to measure low-energy photons in the presence of high-energy photons is greatly en- hanced., An anticoincident spectrometry system has been used successfully 157 20 by Holm et al. in scanning reactor fuel elements and would have compar- able merit in studies similar to our MSRE measurements. The second suggested change in the spectrometry system pertains to an improvement in the accumulation of spectra under transient conditions. Dur- ing transient conditions, such as reactor startup, scram, or modifications of fuel or salt chemistry, it is desirable to collect as much spectral data as possible. To obtain suitable resolution of the large number of peaks and permit computer processing of spectra, it is necessary to collect spec- tra with a gain of about 0.7 keV or less per channel. However, with such a gain a 4096-channel spectrum will only extend to about 3000 keV. Because several of the short-lived fission products emit photons above 3000 keV, their photopeaks will be missed. Under transient conditions there is not enough time to obtain spectra at two different gain settings, and the higher energy portion of the spectrum is lost. The total spectrum can be obtained in two ways. First, a spectrometer system with about 7500 channels could be used; this would permit measurement of the 5230-keV photon of °°Rb, which is the highest energy photon among the fission products with half- lives long enough to permit measurement in most experiments. Second, two spectrometers, one with 4096 channels and another with 400 channels, could be used. Because fission products emit only a few photons above 3000 keV, it is unnecessary to collect spectra at low gains above this energy. The two-spectrometer systems would be used by routing the signal from the de- tector preamplifier to two amplifiers. One amplifier would be operated at a gain of 0.75 keV per chamnel and its output processed by the 4096~ channel analyzer to produce a spectrum with an energy range of 0 to 3000 keV, The other amplifier fiould be operated at about 5 keV per channel and its signal processed by the other analyzer. A zero-suppression amplifier would be interposed between the main amplifier and the 400-channel analyzer to suppress the lower 3000 keV. Pulses above 3000 keV would be processed to yield a spectrum between 3000 and 5000 keV. With either of the above schemes, spectra could be obtained over the full energy range of fission product photons, and no information would be lost during critical moments of transient reactor conditions. 20D, M. Holm et al., "Examination of Fast Reactor Fuel Elements with a Ge(Li) Anticoincidence Gamma-Ray Spectrometer," pp. 22842 in Proceedings of the 16th Conference on Remote Systems Technology, Idaho Falls, Idaho, Mar. 11—13, 1969. 159 APPENDICES 161 Appendix A TABLE OF RADIONUCLIDES The successful analysis of the gamma-~ray spectra, in particular the identification of the nuclides, depends very much on the nuclear data stored in the nuclide library of the spectrum analysis program. Considerable effort was expended in preparing a table of radionuclides for this library, based on the most updated information from the Nuclear Data Project located at ORNL. Since the absolute nuclide concentrations calculated by the com- puter program depend on the supplied nuclear data, we believed it to be use- ful to include this table. 1In case better nuclear data that are significantly different become available, the results presented in Chapter 7 should be changed accordingly. LIST OF INPUT CATA THE HEADING SYMBOLS A,B,C,D,E.F,G,AND H FOR THE INPUT TABLE HAVE THE A=PRECURSOR S5YMBOL MAINLY USED FCR THERMAL NEUTRTN REACTICNS E=PRECURSCR MASS NO. C=PRECURSCR NATURAL ABUNCANCE D=NEUTRON REACTICN CROSS SECTION E= RACIGONUCLIDE TYPE,WHERE C=N,GAMMA PRCDUCT 1=FISSICN PRCDUCT 2=B0TH N,GAMNMA AND FISSION PRCDUCT 3=NEUTRON DEFICIENT NUCLIDE 4=NATURAL RACIUNUCLIDE S=(THER RADICNUCLIDES = N0« COF GAMMA ENERGIES TABULATED G = BRANCHIIG RATIC STATUS,WHERE A=ABSGLUTE R=RELATIVE D=BASED ON DECAY SCHEME GIVEN IN REFERENCE n FOLLOWING MEANING 291 TELE. AT, CECAY NCe ISCTCGPE NOs HALFLIFE UNIT MULOE A B C D E FISSION YIELD 1 019 8 26,500 S B- 0 18 C.2C4 0.00C O 0.0 ket 43 ADAMS AND DAMS J,RAGCICANAL CHEM =-ENERGIES BLUE BOOK FOR OTHER DATA ENERGY INTENSITY G ENERGY INTENSITY G 167.4C0 CT.C00 1356.000 5¢.000 < F 18 2 109.800 M B+ G O.C C.0 3 0.0 REF o ENERGY INTENSITY G 51lieC00 193,400 3 F 2G 9 11.560 S 8- F 15 i00.CcCO C.010 ¢ 0.0 REF 43 ADAMS AND DAMS J,RACIOANAL CHEM —-ENERGIES BLUE BOOK FOR QOTHER DATA ENERGY INTENSITY G 1€232,100 160.000 4 CkR 51 24 27720 D EC CR 50 4.310 17.000 © 0.0 REF 0 NUCLEAR DAYA TABLES SEC A 1970 ENERGY INTENSITY G 220,100 ¢.900 s MN 5¢ 2% 314,000 O EC FE 54 5«.2B0 C.560 © 0.0 REF 36 MARTON, NUCLEAR DATA A4,301 (1968} ENERGY INTENSETY G £z4.810 100.000 € €O %6 27 77.300 D ECB+ C G.C 0.0 3 0.0 REF 0 NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTE 511.G06 484000 733,700 0.110 788.0C0 0.400 846,750 ©77.5C0 1.400 1937.900 12,900 1141,20C C.170 1175.,130 12384300 TC.000 1360.250 4300 1771430 15,600 1811.,000 19£4.200 0.720 2015.36C 24900 2C34,%2C 7.400 2112.800 2213,109 242G0 2274.000 C.120 2374.00C C.150 2598.570 3{16G.,000 0.850 3202.190 3,100 3253.64C T« 6C0 32734190 2451 .400 0.780 3548,200 0.180 7 CC &0 27 5260 Y B~ CO 5% 100.,GC0 37,5060 ¢ 2.0 REF 0 NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G ENERGY INTENSITY G 1173.23¢0 92.880 1332.51¢ 1C0,000 ° 26 A NSITY G 99.974 2.000 0.600 0.370 16.8C0 1,500 €91 E Kk &7 3é 76.000 M B- KR 8¢ 17.370 0.060 2 0.250E 01 13 D REF 1 H LYCKLAMA, ET AL CAN Jo PHYSICS 47,393(1969) ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 403.000 59,700 6744300 2.000 836.000 0. 760 845,800 8,200 1175.,500 1,300 1338.000 0.£600 12384.,00C C.600 1741.,000 2,000 2012.000 2.900 25564000 9.5C0 2559.000 5.100 2881.200 04300 3209.38C0 C.600 < KR 88 3£ 2.800 H B~ 0 0.0 0.0 1 0.360E 01 40 A REF C LYCKLAMA KENNETT CAN J OF PHYSICS 48 P753 (1970} ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G £84G00 0.0 166.000 6.800 196.100 37.800 2404400 0.300 362.L00 3.000 390.400 0.600 4724300 0.600 789.000 0.400 634,700 13.000 862.400 0. 530 B84,.,500 0.070 945,200 0.260 562.200 0.070 986,700 1.600 1017.60CC C.110 1039.600 0.400 1C49,500 G.100 1090,400 c.070 11414700 1700 = 1179.,500 C. 750 1185.100 0,570 1213.000 0.200 124546040 C. 300 1250.000 l1.100 1352,500 0.200 14C6.900 0.200 1518.,500 1,500 1529,800 11.300 1603,829 0.200 1909.100 C.110 2029.50C 4. 800 2035.,300 4.800 1864800 0.150 2195.900 14,900 2231.600 3.600 2352.,400 0.200 2392.000 37.800 2409.,400 0.070 2549, 00C C.260 2771.800 0,040 1C KR B9 3¢ 3,200 M B- o 0.0 0.0 1 0.459E 01 91 A RBS89 REF 2 KITCHING JOHNS NUCL. PHYSICS A98,337(1967) ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 85.600 C.500 93.600 C.400C 150.800 1.000 220,600 25.000 64200 04900 345.300 2.00¢C 356.3C0 €« 5C0 368,800 24000 Z96.000 l.5CO 411,400 1.500 434,500 0.500 439,200 l.200 455,000 1.20C 468, 70C C.750 497.800 li.CCC 527000 0.50C £77.200 2,000 586. 400 21.0G00 €13.000 C.5CC 627,009 1.5C0 £95.0C0 2.000 708,000 0.700 737.60C 44000 Ta44e 000 0« 3C0 7604000 0.5020 777.0C0 0.600 802.000 0500 823.000 2.00G E6U.00D Js70C 867. 500 6.0C0 SC3.50C T+300 $714000 Ce400 G87.700 2. 500 1010.000 1.000 1077.00C 0. 90C 1105.300 5.400 11164500 24500 1173.000 1,000 1273.C00 C.2CC 1298.,00Q0 0.50C 1224.50C 1.600 1370.000 2.800 1472.100 S.5C0 1500.0CC 0.800 1833.4C0 11.000 16364000 1.000 16654000 Ce 700 1670.000 1.000 1€91,68) 4,700 1760.,000 3.00C 1775.000 2.80C 1843,00C l1.1¢¢C 1502.000 1.200 1998,000 0.400 2011.000 2e6C0 2020.,000 170G 2120.C00 1,0C0 2120.,0060 1.700 2281.0C0 2.000 238C.000 Ge4CO 2€18.060C 1.4C0 2644,000 0.80C 2753.0600 C. 700 2T62.C00 0e500 27904700 14400 2865.,70C 24800 2646,000 C.300 2125.0C3 Cel60 2125.000 D160 21432,000 0,500 3219.00¢C C.380 23204000 Ce.250 2353,000 1,200 3284.000 0.259 2480,000 C.200 3510.000 C.3C0 2534,000 1.540 3568.00C CeE50 3629.000 C.080 26532.C00 Q.070 27234C00 7800 2734, 000 0.200 3823.000 0. 080 3834,003 C.080 3842.000 7.080 3894.000 0.140 2904.00C0 C.140 3924.000 0.300 2662,00N 7160 4005, 000 N0.CT0 4040.CO0 C.Cé0 w9t 1 KE SO REF < EMERGY 1215006 a5, 60N 954 .500 SCl.1Cd 1423,.7u) 1£52.100 1824 ,7C 0 cIs82.000 le KB 8E REF 1 cNERGY £€8.0L0 1779.80% 26TT4BCGE 3487.800 13 R3 BR REF 3 ENERGY £C8,.,04 0 1836,133 2734,100 47454297 14 kB 89 REF 10 ENERGY £T7T7.000 5484500 1526,0CGC 2708,000C 1< RB <0 REF £ ENERGY £37.000 10204000 170%.0C9D 2720002 3540,0C0 ACOO.("” 35 22.0C0 S B~ C G.0 C.C ] 0.560FE 21 31 A KB=-S0 JOHNS ETAL BULLAM PHYSLSOL. 12,N05,6¢7 (1967} INTENSTITY G ENERGY INTENSITY G ENERGY INTENSTITY G ENERGY INTENSITY G 58,000 192.008 C.200 234,100 44600 241,800 17.06¢C 1. 0CN 434.100 2.900 455,500 24500 525.8(C0 38.0C0 H4 50N 7321.100 1,700 7Bt 10C €. 400 941.900 1.600 T.2C0 1118.70C 52.000C 1332.000 C. 700 1310.0GC 0.600 2,70 1463,500 0.509 1201.200 0.0 1537.700 13,000 Z.10C 16584200 1.¢GC 1714,2C0 0.040 178C.000 Ge400 Ce4CO 2378,000 0.500 27C8.000C C.200 27264000 1.000 Ne 60N 2853.000 l.0C00 2112.000 l.4CC 37 18,000 M B- kg 87 27.85C 0.120 2 0.260E 01 1= © H LYCKLAMA, ET AL CAN Jo PHYSICS 47,393(196%) INTENSITY G ENERGY INTENSITY G ENERGY INTENSTITY G ENERGY INTENSITY G 18,100 1383,500 1.100 1518.5C0 0.200 1535.,50C 0.200 e300 1836.100 30,200 2115,500C CeS0C 2577.800 0. 206 Z2.90C 2734,100 C.200 3005.500 C.500 2220.000 0.4G0 Le20C 47444500 0.500 4854,000 0.03C 3 17.800 M B~ RB 87 2T.8EC c.120 2 G.36CE 01 14 A NCONE RAGAINI KNIGHT NUCL.PHYS.A125,97 (19€9) INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 14.400 1383,.,500 0.860 1779.40C 0.190 17399,000 0.120 23,200 2118.000 0.560 2577900 Celé0 2677.0C0 2+8C0 G072 3009, 100 C.230 2219.400 C.230 1488, 200 O«140 0150 4853,699 0.00Rr 37 14,900 M B- 0 C.0 0.C 1 0.480E 01 14 A SR=%0 KITCHINGy JOHNS CANeJoPHYSICS VOL 44 P2&61 (1966} USING AU-198,CS-137+NA~24 INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 2.200 250.000 0.500 €58, 800 1C.7C0 770.300 C.400 17,000 1030. 700 £C.000 1225.00C C.720 12464400 464600 3,107 2000.000 4.500 2194,500 17,100 2567.000 12.000 2,000 22C0.000 2.700 37 2«90 M B~ 0 0.0 0.0 1 0.581E 01 232 R SR9Q JCHNSONZO'KELLEY EICHLER PHYSLREVse 1354B836(19¢4) INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 54600 720,000 €.500 832.000 1C0.000 E6C. 000 100.000 8.500 1110.C00 13,209 1240.,00C 4,2C0 1432.000 84300 $.100 1820, 000 4,8CC 22C0.0C0 2,600 2510.2€C0 3.400 l.8¢CC 2840.,00C 2.0000 3070.000 Ge7CO 3240.000 27.500 8.110 4120,000 19.600 4340.00C 31.200 4270.000 21.300 BaallC 5080.C00 2,060 £220.0CC 6.700 9T le SR ¢l 38 S.700 H B~ 0 0.0 0.0 1 0.581E REF 42 GUNNINK ETAL UCID 15439 14JAN 1969 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 2754450 1.33C 6204170 1,900 631,360 04610 6524 980 749,770 24.00¢ 761,300 C. €00 $25.77¢C 4,000 1024.250 12604540 0.940 1413,400 1.400 17 Sk 92 38 2.700 H B- c 0.0 0.0 1 0.530E REF 42 GUNNINK ETAL UCID 15439 14JAN 1969 ENERGY INTENSITY G 12864000 90,000 18 Yy °0M 39 3,100 H IT Y 86 100.000 0,001 O 0.0 REF 43 ADAMS ANC DAMS JL.RADIOANAL CHEM —ENERGIES BLUE BOOK FOR OTHER ENERGY INTENSITY G ENERGY INTENSITY G 2024400 97.000 479,200 91.000 15 Y %0 38 64,000 H B~ Y 86 1€0.0C0 1.300 2 0.577F REF 5 NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G 1742.700 C.016 0 Y 91M 39 50.000 M IT 0 0.0 0.0 1. 0.340E REF 42 GUNNINK ETAL UCIC 15439 14JAN 1969 ENERGY INTENSTTY G £55,630 £3,00C 21 y o1 30 59,000 D B- ¢ 0.C 0.0 1 0.590€ KEF 42 GUNNINK ETAL UCID 15439 14JAN 1966 ENERGY INTENSITY G 12084000 0.220 2z Y 62 35 3,530 H 8- 0 0eC 0.0 1 0.590F REF 42 GUNNINK ETAL UCID 15429 14JAN 1569 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 448,107 2.70¢C 492,100 C.600 560,92C 24660 844,300 C34.E23 14,000 1405,200 4,800 23 Y 93 25 10.100 H B- 0 040 0.0 1 0.610F REF 42 GUNNINK ETAL UCIC 15439 14JAN 1966 ENEKGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 267,059 64400 478,400 04011 68C. 250 0. 700 £95,700 S47,070 2.200 1158.400 0.037 1162.8C0 Co044 1183,700 12044200 6,320 1237,700 0,023 14064600 Ce051 1425.460 1450.87¢C 0,342 1470.300 0,055 1642.5C0 0.051 1917.,850 134,800 Ce120 21G61.20C Cel7C 01 10 A INTENSITY G 12.000 23.000 01 i A 2 A DATA 01 1 A o1 1 A 01 1 A 1 & A INTENSITY 6 1.750 01 1§ A INTENSITY G 0.27C De044 0.237 1.450 91 92 991 ca kR ©5 49 €5.000 D B- IR G4 17.40C0 0.C080 2 Qe £Z29E R:F 7 LEGRAND ETAL,s NUCL4PHYSe A107,177 (1968} ENERGY INTENSITY G ENERGY INTENSITY G T244240 43,600 7564870 54.E80C & IR 97 40 17.000 H @~ IR Q€ 2. £0C g.a0s8¢c 2 0.5G0E REF “2 GUNNINK ETAL UCID 15439 14JAN 1966% ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 254.2C0 1.2C0 3%5.700C 2.200 602.5C7 14280 703.8C0 743,279 24.0C7 1148.000 282G 1276.100 G. 300 1262.7C0 17504609 1.080 1852.000 0260 e NB S5M 41 90,000 H IT IR 94 17.4CG 0.0860 2 N.128E KEF 43 ADAMS AND DAMS J,RADIOANAL CHEM -ENERGIES BLUE BGOK FOR DOTHER ENERGY INTENSITY G ¢35.7C0 107,.,0¢C0 27 NB 95 41 35.150 D B- 0 0.0 0.C 1 O.641E ReF 1 LEGRAND ETAL, NUCL.PHYS. ALQT7,177 (1968} ENERGY INTENSITY G TéE.84C 100,000 8 NB S€ 41 23,000 H B- 0 C.0 0.0 1 D.600E- REF 8 NONARC+BARRETTE,BOUTARD CANeJoPHYS. 46, 2375(1968) ENERGY INTENSTITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 2184900 2.900 241.600 3.300C 350.200 1. €00 352,200 271.800 2,900 460,100 26.800 4804500 5.6G0 568,700 $91.70Q 1,000 719.700 6.700 721.700 1.200 778,200 £10.,4G0 9.000 812,500 5.50C 849,800 21.000 1091.,400 1128.,000 C,700 1200.100 2C. 509 1441,300 0.4C0 1497,00 1497.%00 0.400 29 NB 9TM 41 1.000 M IY IR 96 2800 0.050 2 0.59%E REF 43 ADAMS AND DAMS JJRACIDANAL CHEM —ENERGIES BLUE BOOK FQOR (OTHER ENERGY INTENSITY G 7424260 98,000 30 NE 97 41 72,000 M B- IR S¢ 2.800 C.050 2 0.620E REF 42 GUNNINK ETAL UCID 15436 14JAN 1969 ENERGY INTENSITY G €58.180 9¢,.200 01 2 01 10 A INTENSITY G 1.000 1.080 Qo 1 A DATA 01 1 02 21 R INTENSITY G 1100 56000 100.000 EC. QGO 2«56G0C 01 1 A DATA 01 1 A NE-3E NE €7 L9T zl MG 99 42 66,690 H B~ MO ¢&¥¢ 23.780 0.150¢ 2 0.610F REF 37 VAN ETJK ETsALs NUCL, PHYS, A121,440(1968) ENERGY INTENSITY G ENERGY INTEMSITY G ENERGY INTENSITY G ENERGY 18.251 11.000 18+.367 0.0 40.584 €+ 3C0 140.511 142.630 0.900 181.060 7.600 249.000 C.005 3664400 280.700 0.010 409,000 0.001 411,500 G« 024 458,000 528.900 0.C5%3 620. 700 0.004 €20.7C0 G.024 739.700 T7€.,200 4.800 823.190 0.140 961.000 C.110 588.200 1€21.700 0.C06 1016.C00 0.001 32 MO1l01 42 14,600 M B- MOl100 2.630 c.2C0 2 0.500E REF l4 CRETU ETAL REV.ROUM,PHYS.,TOME11l,N0O2,P143 (1966) ENERGY INTENSITY G ENERGY INTEANSITY G ENERGY INTENSITY G ENERGY BC.J00 2,000 190.000 T.000 191.0CC £9.000 300.000 40G.000 2.060 510.000 2.000 51C. 000 18.000 590,000 700.000 14.0C0 890.000 14.00C0 950. 000 1.0C0 1020.000 1180.000 9.000 1280.000 2.709 1380.000C G.000 14460.,000 156u. 000 12,000 1680.000 4,000 1800.,00C 1.0C0 1890.000 208C.000 9.009 2160.600 1.000 23 TC 9SM 43 6,000 H IV MO 98 23,780 0.510 C 0.616E REF 0 NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G ENERGY INTENSITY G 140.511 89.500 142. 630 Q.027 e TCl01 43 14,000 M B- MO100 S4620 0.2C0 2 0.502E KEF 43 ADAMS AND DAMS J.RADICANAL CHEM —-ENERGIES BLUE BCGK FOR OTHER ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 127,300 3,000 306. 800 91.000 544,500 8.000 626.£00 2 TC10Z 432 4,5¢C M B~ 0 0.0 0.0 1 O0.41FE REF ¢ BLACHCT ETAL MNUCL PHYS Al13¢ (196¢) P434 GE DIODE eNERGY INTENSITY G ENERGY INTENSITY G ENERGY INTERSITY G ENERGY 415.300 2.0N0 4184800 3,400 474,800 160,000 627.500 £30L.E00 29,700 691,800 2.700 695,600 3.000 504.000 G21.900 0.0 10464200 12.800 1073.00¢C 1.000 1104,£20 11154000 0.7 1128.0C0 0.0 1197.80¢C 10.8C0 1292.6C0 1221.80v 3,700 1811.C00 24500 1595.7C0 £+100 1612.,800 1€17.060 11.1C0 1710.000 5.500 1810,00¢C T.000 1907.000 is5C.000 0.0 2141.600 1.C0¢C 2227.C00 7«100 2242.2C0 233%,200 10.060 242 7.800 c.C 3e TC10Z 43 5.000 S B- 0 0.0 0.0 ¢ Ds415E REF 0 BLACHOT ETAL NUCL PHYS Al3G (196¢) P4324 GE DIOCCE ENERGY INYENSTITY G ENERGY INTENSITY G ENERGY INTENSITY 5 ENERGY “4E8 4000 15.000C 4TE.000 1CC.000 £28.000 15.000 1102.€00 11¢e.000 3.000 01 22 A INTENSITY G 95.100 1.450 0,005 13.760 0.002 01 22 D INTENSITY G 64000 25.000 25000 3.0C0 1.0C0 01 4 A DATA INTENSITY G l1.000 01 30 R INTENSITY G 17.000 .0 12,200 5. ‘006 T« 300 00 13.5C0 01 5 R INTENSITY G 11.€00 TCO99M TC-101 43 99 g9l =7 PULG2 44 2%.6(0 D B~ grUl02 314610 le4CC 2 C.2C0E @1 7T A rEF ~ NUCLEAE DATA TABLES SEC A 1970 tNeR oY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 2%.559 D.08% 524110 Ce340 294,880 Calll “43,850 Ce3£C L9 . 90T £9.0CC 556,900 0.8CC €10, 200 400 b RULCS 44 4e430 H F- RUL104 16,580 C.480 2 0.5C0GE €O 17 A REF &2 GUNNINK ETAL UCID 1542¢ 14JAN 1S5¢5 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 1254530 4,800 149.040 1.580 2624840 €.550 2164502 10,200 32€4100 1.070 3304500 D.72F 2FC.18C 1.220 393.32€0 2.8¢€C “134E1% 2.200 46943280 17.5CC 499,289 24270 575.190 1.020 €EEo.154 1.920 6764320 18,0082 724,20C 44,5C0 875.8C0 24940 26T 4,400 2.000 g FPH102M 45 £7.0C0 M IT RU102 31,610 le4CO 2 0.0 1 A kEF 0 NUCLEAF DATA TABLES SEC A 1970 ENERGY INTENSITY G 364550 C.MNiBE <0 RH1GSM 45 45,000 S IT C 0.C 0.0 1 0.850E QO 1 D KEF 17 LEDERER,HOLLANDER,PERLMAN TABLE OF ISOTOPES (1%568) eNERGY INTENSITY G 125.430 100,200 41 RH105 45 1.475 D B- RULO4 l18.580 Ce480 2 0.900€E 00 6 A REF 4z GUNNINK ETAL UCIC 15439 14JAN 1966 ENERGY INTENSIYY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY & 212.€E00 0.113 280,540 0.180 306.310C Ee44C 319.240 194600 442.20C 0.037 427,000 0.05¢ LE107 RH105 45103 69T 42 FH106 45 367,000 D B- 0 0.C 0.0 1 0.2S0E 00 106 A REF 13 STRUTZ,STRUTZ,AND HAMMERSFELD, ZePHYSIK 221 231(1969) ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 22B.3200 0.000 427.900 0.092 435.C00 0,047 498,500 0004 511.800 2045C0 £44,C00 0.C03 £5249C0 0.C02 565,700 0.001 56G.200 04002 578,000 C.008 588.100 C.001 602.500 0,002 £1€£.200 0.810 621.800 9. TE0Q 6614260 C.015 680.100 G.007 684.30C 0.002 7164200 0.016 717.100 0.01% 749,500 0.002 774.600 0.006 873,1C0 0.414 1045, 70C 0.004 1056.,100 1,450 1062.100 0.024 1108.900 0.C05 1114.¢0C0 0.008 1128.000 . 0.383 11504100 0,000 1159.700 0.000 1179,100 0.014 1180.600 0,014 1194.300 01.053 1220. 400 0.000 1231.700 0.001 1257.000 0.001 126€.c00 0.001 13C5.6C0 C.003 1310.700C 0.0C3 1316.800 0.004 1355,000 0.001 1359,900 0.001 1397,.300 C.003 1457.800 0.001 14688, 5C0C 0.002 14964400 0.026 1562 .C00 Cal46 1574.400 0,001 12T7€.£20 0.002 1592, 700 0.008 16007060 C.008 160G5.700 0.001 1609.800C C.000 1686.100 0.001 1730.000 C.001 1733.100 0.002 1766.,200 0.023 1775,000 0.001 1784, €00 C. 00 1789.000 0.006 1796.70C 0.02% 1854,800 o.00¢® 19044000 0.000 1909.400 0.006 1926.8C0 7,013 1927, 200 0.0113 15504400 0.001 1956.900 G.001 1$88.500 D.027¢ 2014.,000 C.000 2033.90C C.0CO 2041.900 Q.000 2064.100 0.006 2169, 200 0.000 2193,200 C.00k 2194.000 C.005 221C.700 0.0C0 2230.400 0.000 22424400 0.001 2271.700 0.001 308,700 G.006 2309,400 C.006 2317.¢€00 0.006 23664400 G.021 22904800 0007 2406.C00 0.012 2438,500 0.004 2485,700 0.001 2£25.200 0N 2543,400 0.002 25714200 C.001 26514600 0.001 27064600 0.0032 2709.400 0.003 2740.,500 G.00C Z2783.500 0.0 £80%.50U 0.004 28214200 0.001 2504.100 €.00¢C 2917.9C0 0.001 2027.100 0.0 3036.8C0 0.001 3055460C C.0CC 3085.,900 0.0 2161.500 0.0 3169,000 C.0 4z KH107 45 4,200 M E- c C.0 0.0 1 C.190EF 0C g PD-107 KEF 12 GRIFFIN,PIERSON, BULL«AMusPHYS4SOCa13,NO.4 (196€8) ENERGY IMTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 112.600 0.0 202.700 0.0 312.200 C.0 348,200 Ne0 Z8l.70C 0.0 39244C0 0.0 56T7«6CC .0 670.000 0.C L4 AGL1ICM 47 260,000 O EB- AG1OS 48.180 2,000 ¢C 3.0 15 A REF 38 BEMIS,ETAL, NUCLe. PHYS Al2% (1969) NO 1 P217 ENEKGY INTENSITY G EMERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G a4€ . T70 3,200 6£204229 24740 6574710 C4,35C ETT7.550 11.510 tBELECD 6o 8N 7064 €R0 16,230 744,190 44150 763.280 224640 Bl8,.0C9 Te?€C 684,670 7.1C0 G37.480 244440 1284.220 2¢.120 1475,730 G250 15C4.200 12.260 15624220 1,250 0LT &5 £Bi25 FEF ¢ ENERGY 35,450 116.379 178,78y 221.120 443,620 £J64 820 46 SB126 REF 42 ENERGY c23.SC0 EEE.20C 20,000 674a9CC £56. 700 “7 $Biz27 KREF “#2 tNERGY 252.70C 2G2.1C0 465,200 603,8C2 £84,900 753,470 216,400 1289.50C «B S$8129 REF 0 ENERGY 18¢.0C0 523.000 760,000 967.000 1200.020 2070.000 4< TEL12TM KEF ¢ ENERGY 58.C02 £l NUCLFAR DATA INTENSITY G e A00 0e260 C.043 Oe44C Ne280 52C0 51 GUNNINK ETAL INTENSITY G 1.7¢0 1.480 1.15C 4e 320 17.600 51 GUNNTNK INTENSITY G 6e7C0 l1.000 4+300 4. 4CC 36.800 D.0r7 0250 04340 51 INTENSITY G 24¢2C0 240090 1.600 5. 7CO 1,600 D800 52 INTENSITY G 0.010 FeGED C.100 INTENSITY G D.068 C.200 C.l9C C.240 C.250 l.800 C.0 INTENSITY G 4,TF0 CL00C 2e830 32.000 £.500 G.C INTENSITY G 1.800 C.780 C.670 C.260 1.800 C.140 C.5CC Cal INTENSITY G 543C0 2.8C0 2900 1,060 4.100 18,710 0.100 3.210E-C1 z3 & ENERGY INTENSITY G 111,060 C.0G8 176,299 €.3CC 227.<00 0.1C0 427,950 294600 6004770 15.400 0.320E-C1 1¢ A ENERGY INTENSITY G 414,7C0 £1.000 £05.000 24480 6664200 100000 T2N.4€0 564000 0.137TE CO 20 A ENERGY INTENSITY G 309.60C 0.210 440.700 CasT0 542,2C0 3.000 6664900 G.64C 7434700 0.150 8064000 C.C36 1140.3C0 C.3280 0.1C0E 01 21 ¢C ENERGY INTENSITY G 412.0C0 2.000 682,000 54300 €16,.,000 17.000 1240,000 2.800 1840, 000 1.000 N.127E 0O 4 A LEDERER, HOLLANDER GAMMA AB INTENSITY G C.3200 2.770 ¥ IN124 TABLES SEC A 1670 ENERGY INTENSITY G ENERGY 81,800 1.00G 109,270 122,430 0.03¢ 172. €GGC 204,070 0.250 208,120 380.¢°10 le4nC 408.1G0 463,510 10.000 489,800 €3€.15¢C 11.200 £T1.560 12.50C D 0 0.C UCID 15429 14JAN 1969 ENERGY INTENSITY G ENERGY 278,00 242850 297200 57T3.70C T«300 5934CC0 639,009 1.350 €56,2C0 695,100 10C.000 587,00C 554,200 12.3200 $89. 500 3.500 O 0 0.0 ETAL UCID 15439 14JAN 19669 ENERGY INTENSITY G ENERGY 28C. 800 Ce4CO 220.60C 412,000 3.500 417.500 473.200 24.E00 501.9C0 638.20¢C 0.500 652. 700 £97,90¢C 24500 72144CC 782.600 5.0€C T87.900 829,100 0.249 924,200 1379,000 0.068 4,350 H ¢ 0.0 YOSHIHIRD TAGISHI JePHYS.SOC.JAPAN 21,2429(156¢€) ENERGY INTENSTTY G ENERGY 296,C00 1.600 3584000 £44,00C 16.0C0 652,000 813,000 4C.S0C 8764 CCO 1028.000 10,000 1048.,000 1520.000 24400 1730.00C 169,000 D TElR2E€ ADAMS ,DAMS J RADICANAL CHEM.3,271(196°)DICDE ENERGY INTENSITY G ENERGY 361,000 0.050 417,400 ENERGY 563,000 INTENSITY G Cel 5zl2% TELZ7 52129 LT 0 TE129 52 76,000 M B~ TEl28 31790 0.140 2 0.900E REF 0 DICKINSON FT.AL, NUCLEAR PHYS A123 481-496(1969) cNERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 27,800 17.660 208,980 0.199 250, 65C C.430 270.300 2784430 0.614 281.160 0.161 3424600 C.008 342,800 459.6030 T.€80 4874390 le460 £31.830 0.062 551.500 559,7C0 1.008 624,400 0.092 7164800 0.002 T41.100 768,500 0.004 794,500 0.002 B02.170 C.209 8294900 833.4C0 0.045 g2%5.800 0.001 982.4C0 0.017 1003,4600 1C12.830 C.,001 10834990 0.606 1111, 740 C.238 1204,200 1233.000 J.009 1254.200 0.001 1260.840 0.012 1264,400 1282.100 N.001 5i TE129M 52 33.000 D B-IT TeEl28 31,750 0.017 2 0.350¢E REF 0 DICKINSON FT.ALs NUCLEAR PHYS A123 4R81-496(198&9) ENERGY INTENSITY G ENEKGY INTENSITY G ENERGY INTENSITY G ENERGY 105.500 C.230 £E564650 0176 €72.G30 c.028 695,980 701.800 0.029 705.600 c.00C8 729.620 1,170 T4l1.100 7684500 D.005 617.200 C.145 844,900 C.05¢ 1022.600 iC5C.«00 Ue028 1373,800 0.001 1401.60C C.0G7 5o TE131M 57 1,20 D 78~ TE13C 344,480 0.030 2 D.440E REF b2 GUNNINK ETAL UCID 15439 L4JAN 1969 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 102.2G0 7.500 149.800 244200 182.4C0 2.100 200.700 240507 84.40C 334.300 11.1C0 4524400 te5C0 665.100 773,720 46.000 7824500 7.€00 753,80C 1%.900 822.800 €52.200 25.600 1125.500 14.8C0 1206.€00 11.8C0 164£.8C0 i687.7CO 1,70C 2001.000 2700 ‘ t= TE131 52 2%.0060 M B- TE13C ®kdkddok*kx 0.2C0 2 De260E REF C ADAMS ,DAMS J RADIOANAL CHEM.3,271(1969)DIODE LEDERER HOLLANDER ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY . ENERGY 149,7C0 69.000 343.000 0.C 384.000 0.0 4524400 +92,70C S.0C0 €02.10¢C 4.000 654.4C0 1.7C0 Q24,C00 €49.009 C.0 ©97.200 4.C0N loc8.C00 0.0 1147.800 Ha TE132 52 78,000 H B- 0 0.0 CaC 1 0.433E REF n NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY 49,7293 1Zz.970 111.760C 1.80C 11643600 1.900 22841690 00 33 D 53129 INTENSITY G 0.005 0039 0.012 0.035 0.008 0.001 0.000 G.009 o 15 D 5312% INTENSITY G 44510 0.045 0.031 Q0 18 A 1131 INTENSITY G T«&00 5.000 6. 700 1.500 01 lz & 52131 GAMMA AR ' INTENSITY G 16,000 0.0 £.000 01 4 A 53132 INTENSITY G ES.000 cLT 5E TE132M 52 55.40C M B- 1 0 G.0 0.0 1 N.65CE 01 £3 v TE-1332 kZF 2C MCTSAAC PHYS.REV VOL172 P1252 4 HEATH IN-1218 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSTITY G ENERGY INTENSITY G T4..00 J.8CC 81.5G0 U.800 88,000 2.000 94,500 £.0C0 164,340 1.600 168,870 12.000 177.10C 1.5C0 178.2C0 1.000C 184,450 Fe400 193,220 l.2CC 138.200 C.600 213456C 44200 22Ue940 0500 2244030 C.400 244,280 Ca70GC 2514460 Ue €00 257.€40 i1.009 261, 550 14,CC0 234,140 1£.000C 344,500 1.500 247,220 1.300 255,570 0.609 262.81C l.10¢C 376.820 0.6C0 296496C 1.70C 429.020 3460G 444 .90C 44400 4624110 2.40C 471,850 2.30C 478+ €90 1.8C0 519.60C C.5C2 534,850 24060 5T4.C40 34600 622.030 1.600 6474400 24,000 7024752 443C0 T3i.590 1.700 T32.89%0 3.300 71¢. 750 2.90C 795,700 1l.50C SuL. 510 2.200 863,910 29.000 982.8307 4. 800 BG7,700 C. 500 Sl2.530 1C0.000 Gl4.720 13,000 S34.40C 1,50C ©78.,199 %200 SEu.4L0 2.7CC 982.900 1,300 1026,.,800 l.800 1061.830 3.100 12438,200 20900 1459,100 2.5C0 1516.100 1100 i531.600 1.06C 1587.450 2ec!0Q 1683, 30¢C 6.7C0C 17C4,400 1.1C0 1855,700 1,300 c004.900 5.400 2027.700 1.400 204G, 200 1.7C0 e TEL133 52 12.400 M c 0.C 0.0 1 0.650E 01 31 A I-123 REF ie PARSA,GOGRDONsWALTERS, NUCL.PHYS.ALl10,€74 (1668) eNERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 212,100 70,000 384, 600 0.400 292.900 C.8C0O 407.900 31.000 475,000 1.100Q 5464400 C.800 587.100 0. 7C0 613.6C0 0.400 €96+500 Do P 7204100 8.300 787,000 T.2C0 844,500 446C0 $3C.580 5.400 1000.5C0 4500 1021.0CC 24400 1¢61.000 1.800 1296.500 G.0 P 1252.1C0 1.4G0 1307.700 1.30C 1313.5C0 1.1G0 1332,300 11.090 1405, 6C0 0.800 1474.,00C0 " Qe.ECO 1518. 600 0.700 1588.200 C.400 1717.570 3.2G0 18254100 G.8CO 1881.450 1.3CC 2i34.500 0.400 2228,000 €400 2540,60C 0.100 57 TE134 52 41,800 M B- 0 0.0 C.0 1 0.4£90F 01 13 A 1-1324 KEF %1 TE-124 C.E. BEMIS,ET.ALe ARK. FYS. 37, 202 (1968) ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 79.500 53.000 101.40C 0.500 181.100 22.000 201.500 9.400 213.800 25,000 278,100 21,000 434,800 18.000 4604700 Be9CO 46k 400 44300 565.600 15.0C0 712,500 £.100 742.000 14,000 766.700 27.000 568 I 131 53 8,060 D B- TEL30 34.450 0.240 2 0.291E 01 15 A 54131 KEF Q NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G ENERGY INTENSITY G 80.164 2.450 163,570 0,023 177.230 C.270 272.300 0.070 284,312 5.8920 318.009 0.090 325.00C 0.033 225.750 C.280 238.500 0.01¢ 3644490 82.40G0 405,000 Ce. 066 5024940 0.330 €36.900 ¢£.9C0 €43.000 0,150 722.51C 1.620 eLT £G 1 132 REF 39 ENERGY 147.200 2104500 4464000 £05.940 £21.000 670.000 780.200 Bl2.290 954,550 1096.800 1143.40G0 12©5,200 12724100 1476.800 1715.,400 1985,500 22234150 &0 1133 KEF 42 ENERGY £t2.500 6174560 £20.500 123€e000 1896.%CG 1134 5460C CRNL~-11C-4 GERRARD:+PINAJIAN,BAKER INTENSITY G ENERGY INTENSITY G ENERGY 04240 255,000 G.150 2844600 0.160 363. €00 C.504 416.800 0+ 680 473.400 0.270 477.960 44960 522. 640 16,400 547,100 2.000 630,210 14,000 650.50C 4400 671,700 6.100 727.000 1200 784.500 0.300 792.100 5700 8764900 1.100 S10.300 17.900 5844500 0.730 1035.20¢C 0.03¢ 1112.500 0,063 1126,.,600 1.4CC 1148,2€0 0.210 1173.300 1.800 1298.000 0.77¢C 1314.300 2s+3CC 13984,570 64800 1442.560 0.130 1542, 200 0.010 1593,100 0.053 1727.,200 0.057 1757.500 N.008 2002,310 1.090 2CB86.840 Ne098 2249.000 0.030 2487T.600 53 20,900 H B- 0 0.0 GUNNINK ETAL UCID 15439 14JAN 1969 INTENSITY G ENERGY INTENSITY G ENERGY 04370 422,800 0.300 $10.530 D560 680,410 0.650 T06.71C N«1320 8564470 1l.220 875.540 1.500 1298.400 24240 53 £3.000 M B~ ¢ 0.0 GUNNINK ETAL UCID 15439 14JAN 1969 INTENSITY G ENERGY INTENSITY G ENERGY 34260 139,100 C.650 162.470 1.740 4054 44C Te3EN 433,300 1.610 514.3E0 2.380 540,800 1C,. 900 628,000 24500 £77e340 3.830 847.030 96,000 857,280 4,000 9744630 4,930 1040,G00 S«150 1455.500 2960 1613.7CC 0.0 INTENSITY G 0.800 0.470 C.100 1.200 24500 64500 €.080 C. 906G C.570 0.052 1.100 0. 060 1.420 04045 C.340 0.23C 0.C02 0.0 INTENSITY G 24000 1.530 4,430 0.0 INTENSITY G 0e260 44450 8.6C0 84200 €.600 2. 000 4,000 1 0s430E ENERGY 306,600 431.900C 487.500 590,900 667,680 772.600 802,800 927.700 10864300 1136.020 1290, 800 1317.200 1456.500 16204600 1921.100 21724670 25254120 0. 669E ENERGY 529.910 T68.5G0 1052.500 0.T780€F ENERGY 128.420 4594000 59544C0 7304600 884,080 1072.530 1741400 01 68 A INTENSITY & 0.110 0.460 0.180 0.060 99.220 75900 2.700 0.410 0.070 2.900 1,100 0.090C 0.050 €.030 1.200 0. 200 0.036 01 14 A INTENSITY G 89.000 0.450 0.495 01 29 A INTENSITY G 0. 680 l.430 11.200 2.20Q 66.000 14,300 3.100 XE133 7LT bz 1158 REF 42 ENERGY 158,190 425,500 5454563 G72.21C 1131.570 12€7.72C 15¢L.600 1832¢,800 63 112¢ REF 22 ENERGY 187.7C0 381,7GQ 1213,.,20¢ 2287.800 2576.800 53 GUNNINK ETAL INTENSITY G Ge 500 0,339 O« 400 8.400 24700 264800 Qs 750 16540 0700 53 LUNCAN,STIVOL A, INTENSITY G 44200 8.2CH0 82.000 7400 1.800 o4 XEL131IM 54 NUCLEAR DATA TABLES SEC A 1970 ENERGY INTENSITY G RE F 0 80.164 2.450 &5 XEL133M 54 REF 0 ENERGY 233,5C¢ ¢é XE133 REF 0 ENERGY 79.550 14.000 54 ENERGY 163.970 2,260 D 5290 D IT Pe ALEXANGER NUCL PHYS INTENSITY G B INTENSITY G 0,023 NUCLEAR DATA TABLES SEC A 1970 INTENSITY G 04040 67 XE13EM 54 REF 0 ENERGY . 526,800 68 XEl35 KEF 23 ENERGY 156,500 IT:a10C 654,630 INTENSITY G 80,060 54 INTENSITY G D262 0.012 De0?25 ENERGY 80.995 15,60C M ADAMS DAMS,J«RADIDANAL.CHEM, 9,100 H ALEXANDER, ENERGY 199,900 408,200 721.9C0 IT B- INTENSITY G 364600 0.020 Ce339 0.0%50 XEl32 A121 612 (1968) XE132 XE1l34 3,271(1969) XEl>4 LAU NUCL «PHYS.A121,612 INTENSETY G 6.700 H B~ ¢ 0.0 UCID 15439 14JAN 1966 ENEPGY INTENSITY G ENERGY 2204400 <¢20C 229.€8C 408,000 O.58C 414.800 433,500 0,600 453.60C 707.900 0.850 769,350 1038.810 1C.0C0 1101.580C 1165.100 1,290 1240.,400 1368€£4900 C.720 1457.,¢10 16784260 11.800 17C6,70C0 1927.300 0370 245,000 83.0C0 S B- c 0.0 ANNe ACAD, SCI. FENNe, ENERGY INTENSITY G ENERGY 219.700 0.600 345,500 295,800 1.000 1246.400 1320.,200 18,009 1767.,200 2414.,000 2.800 2631.800 32039.500 1.200 3153.70C 11,900 O IT XELr30 ENERGY 1€0.5Q0 (1968 ENERGY 249,650 573.200 812.€00 C.0 INTENSITY G 0.250 C.3¢0 Ce320 C.0 P 1.8C0 1.020 1C.CCC 4.50C C.900 G.C SER. A VI,NO. INTENSITY G Z2.100 34300 l.4C0 4.20G 1,200 4,080 £.000 26.890 0.022 2€.890 0.200 INTENSITY G 0.051 10440 0.003 104440 54200 INTENSITY G G2.0C00 0.055 0.617E C1 38 A ENERGY INTENSITY G 288, 3€0 44 00C 417,660 4. 1CC 526.540 1£,409 836,880 8.000 1124.000 49200 12604500 34,900 1502.800 14240 1791.400 Se4C0 0.640E 01 20 R 2€88,1-15{(1368). ENERGY INTENSITY G 371,200 241G0 12568.600 Te 200 1¢35,0C0 1.4C0 28564500 24700 3238.000 1.700 0e22CE~01 2 A 0.160E €O 1 a D.6€69E 01 3 A 0.180F Q1 1 A 0.6405E 01 12 A ENERGY INTENSITY G 258, 600 0e239 €084 600 20630 1043,000 0.003 XE135 XE133 54135 €S5-135 GLT 69 XEL137 REF 2% ENERGY 394,000 849,000 969.000 1117.000 1230.000 1657.000 15164000 2E52.000 3914.,000 70 XE128 REF 28 ENERGY 153,200 401.600 2013.000 71 XE136 REF 29 ENERGY 121.000 R i . | | | [ i | | _8 CASE NO. COLLIMATOR SHIELDING MATERIAL 10 DIAMETER {in.} NONE > {in, Al Yoin.Cu + 5 in. Al ; Yoin.Cu+1in. Al 2 e NONE Ygin. STEEL "/ain. Cd -9 . R 10 Yg in. STEEL + g in. Cd 5 100 300 500 700 900 1100 1300 1500 1700 {900 2100 2300 2500 2700 2900 PHOTON ENERGY (kev) Fig. C.3. Absolute efficiency of gamma-ray detection system of primary heat exchanger. €61 194 Appendix D CALCULATIONS OF COUNTING EFFICIENCIES FROM FIRST PRINCIPLES Objective We have computed from first principles counting efficiencies for the MSRE main off-gas line (see Sect., 5.4) and the MSRE heat exchanger (see Sect. 5.5). Initially we believed that such calculations could be done by models that were of necessity so simplified that only gross agreement be- tween theory and experiment would be achieved. However, this has not been the case; very close agreement has been found over a wide gamma-energy range, Although significant differences still exist below 200 keV and above 1500 keV (precbably still due to the simplified calculational model), the calculations serve to confirm the validity of the empirical calibratiouns, MSRE Main O0ff-Gas Line Since the MSRE main off-gas line was only 1 in. in diameter we used a line source as a model in our calculations. The physical configuration of the detector-collimator-source is shown in Fig. D,l1. Distances of items labeled in the figure are as follows: Lco = geometrical collimator length D = collimator diameter = 1/8 in. o S 12 in. = 30.48 cm, 0.317 cm, fi = distance from source to source face of collimator = 15 ft = 457 cm , L length of source subtended by collimator. Due to penetration of the edges of the collimator on both source and de- tector ends, the length of the source subtended by the collimator is not constant but depends on the gamma-ray energy. Mather,' in a study of col- limators found that, to a first approximation, the effect of edge penetra- tion is to reduce the length of the collimator by two mean free paths of the gamma photons. The effective collimator length Lc thus can be computed by 'R. L. Mather, J. Appl. Phys. 28, 1200 (1957), 195 L co ‘(30.48 in, ™™ r/\N\ Dc = 1/8 in. __SOURCE- COLLIMATOR DETECTOR — Y 2 457 .2 N e— e Fig. D.1l. Schematic of detector——collimator—line source arrangement. LC = LCO—Z/Ut s (1) where My is the linear attenuation coefficient for the collimator material (lead). The subtended source length can then be obtained from 2D L = ¢ £ L - Lc (H + 2) . (2) Thus, in our experimental arrangement, the source length subtended varied from 3.8 in. for 200-keV photons to 4.4 in. for 2000-keV photons. Let S, = source strength per inch = 1 gamma in.~* sec~! and Ac = geometrical area of collimator hole, 7 (0.317)% cn® = 0.07197 cn’. Then if we consider a differential source length, dx, at point x on the source, the photon intensity dG due to dx at the distance 7 + Lco (source- to-detector distance) is given by dq = —— (3) Gn(H + L )? co If the coliimator were not between the source and the detector, the photon current (due to dG) that would hit the detector over an area Ac would be dIl = A4 _dG . e However, since the collimator is present, only those photons emitted from the central portion of the source can go through any portion of the de- tector end of the collimator hole. Photons emitted by parts of the source that are off the source~collimator axis are partially shadowed from the detector by the collimator and can only go through a fraction of the area Ac' Thus Eq. (2) must be multiplied by the fraction of the area 4 that is not shadowed by the collimator. Let this fraction be F(x), and we have dT = 4 R@) d6 , (4) where R(x) is a function of the distance x that the segment dx lies from the collimator axis. Now the quantity A(x) is a radial weighting function and corresponds to the R(x) of Section 5.3. Although R(x) is a gaussian-type function, we can approximate it by a linear function that has a value of unity at the center of the source and a value of zero at a distance L/2 from the axis. Within these boundary values, RF(x) is given by B(x) = 1 — 2 f.’L’ . (5) Thus by combining Egqs. (3), (4), and (5), we have an expression for the photon current dI that leaves the segment dx and hits the detector, viz., 197 2 S, A (1 —=ux) dx a7 = e’ L : (6) 4r(H + L )* co We can now integrate this expression and obtain an equation for the total photon current that hits the detector: o 51 4, L/2 ) I = ————— 2/ (1 —-% ) doe (N 2 0 L 4n(H + L ) coO S, A, L 2 8n(H + Lco) Since the photon current through an area equal to Ac without the collimator being present is SZ Ac L 2 4 (H + LCO) 3 we see that collimator shadowing for a line source allows only one-half of the photons directed toward the collimator exit hole (detector end of collimator) to be transmitted through the collimator. The fraction of photons P' that interact with the detector is given by P' = 1 — exp (-uGe DGe Yy , (8) where is mass attenuation coefficient of germanium (cm®/g), DGe is Ge density of germanium (5.33 g/cm®), and Y is diameter of the germanium de- tector (2.7 cm). Thus the total count rate in the detector is given by CR = IP' . (9) 198 To obtain the count rate in the photopeak, we must multiply by the fraction of the counts that fall in the photopeak. Because these peak-to-total ratios are difficult to calculate, we used values that were measured with our collimated detector with the sources ®'Cr, *®7Cs, *?v, and 2®Al. These sources each emit single gamma rays with energies of 320, 662, 1434 and 1778 keV respectively. The corresponding peak-to-total ratios found were 0,195, 0.106, 0.0551, and 0.046. When these peak-to-total values were plotted versus energy on log-log coordinate paper, a straight line was ob- tained that could be represented by P = 26.828 0:822L (10) where P is the peak-to-total ratio for photons with energy E. Now, combining Egs. (2), (7), (9), and (10), we have for the count rate, 2D L r q 21 & - ( -0.8521 5, 4, [ T 3 J [1 exp(-u,, Dy, ¥)| (26.82F ) CR = - . 8r(H + L )*? co (11) Since the counting efficiency EF is simply G L and SZ is assumed to be 1 gamma in.”* sec ', the right side of Eq. (11) is also equal to the counting efficiency. A computer program was written to evaluate Eq. (11) and obtain count- ing efficiencies at several energies over a range from 100 to 2700 keV. The program allowed for photon transmission through the edges of the col- limator according to Egs. (1) and (2). Results of the calculated effici- encies are shown in Table D.1l, along with experimental values. As can be 199 Table D.1. Calculated versus experimental counting efficiencies for MSRE off-gas line Efficiencies (x 10°) Energy (keV) Measured Calculated Percent difference® 100 15.0 27.2 -81.3 200 12.1 13.7 -13.2 300 8.4 9.5 -13.1 400 6.2 6.2 0 500 5.1 5.1 0 600 4,2 4,1 2.4 700 3.6 3.4 5.5 800 3.2 3.0 6.2 900 2.9 2.6 10.3 1100 2.4 2,1 12.5 1300 2.1 1.8 14.3 1500 1.9 1.5 21.0 1700 1.7 1.3 23.5 2000 1.5 1.1 26.6 2700 1.04 0.79 24.0 aPercent diff. = 100(meas. — calc.)/meas. 200 seen from the table, agreement between calculated and measured efficiencies is good from 200 to 1300 keV; below 200 keV the difference becomes very large. MSRE Heat Exchanger The MSRE heat exchanger® is composed of a 16 3/4=-in.-0D shell that encloses a 14,5-in. diam tube bundle. The shell is made of Hastelloy N and is 1/2 in. thick; the tube bundle, also made of Hastelloy N, consists of 156 cooling U-tubes (312 tubes) plus twenty-three 1/2-in. dummy rods. A typical cross section also contains eight 7/16-in, baffle-plate spacer rods. Although these tubes and rods coated with fission products represent a complex radioactive source, we have calculated counting efficiencies for the heat exchanger using a model in which the tubes and their radioactive coatings are assumed to be homogeneously dispersed throughout the tube bundle. The detector—collimatecr—heat exchanger arrangement is schematically represented in Fig. D.2. The source region subtended in the heat exchanger is a truncated cone with a front face diameter of 3.9 in. and a back face diameter of 4.3 in, As in the case of the line source, because of trans- mission of photons through the collimator edges around the collimator hole, the diameter of the subtended area increases with increasing energy. This effect is actually larger in the heat exchanger because of attenuation in the heat exchanger tubes and shell. However, we will neglect the latter effect and only acccunt for collimator edge transmission in the same manner as before. Thus, the diameter of the subtended area will be determined according to Eqs. (1) and (2). In addition, we will approximate the trun- cated cone by a cylinder with a diameter equal to that of the front face (detector end) of the cone. This approximation is reasonable since most of the radiation measured comes from the front portions of the cone. See R. C. Robertson, MSERE Design and Operations Report. Part I. Description of Reactor Desigr, ORNL-IM~728, pp. 162~72, for a more com- plete description of the heat exchanger, 201 i L" 1/2 in. i I 12 inepee— 15 ¢ - Pb —_— Collimator S S | (Hole Diam = 1/8 in.) Heat / Detector Exchanger x—%J i 14 ins Fig. D.2, Detector-collimator assembly shown subtending a cone~shaped source region in heat exchanger. Because the collimator partially shadows all but the central portion of the source from the detector, we must again obtain an average radial weighting function for the radiation transmitted through the collimator. Because this function is dependent only on the distance from the source- collimator axis (and not on depth of the source), we can derive the func- tion for a two-dimensional, circular area source. Consider the front face of the cone with a differential area, o dp df, located a distance p from the source-collimator axis. The photon current dI through the collimator due to this area is given by A48, ar = —CS2 R o do d8 , (12) 2 4w (H + Lco) where SA is the source strength per unit area and R(p) is the radial weight- ing function. Other terms were previously defined. We approximate F(p) by a linear function that is unity at the source center and zero at the circumference. Thus 202 P Rp) = 1—775 (13) L/2 ° Substituting this result into Eq. (12) and integrating, we have for the total photon current: Ac S L/2 27 0 — —_—_—-.—-——.—.—.-.—-A - ——— ro- 2l [ =gz e do do (14) 0 4 (H + LCO) 4, 5y m(L/2)2 2 4r(H + Lco) 3 The photon current through an equivalent area Ac without the collimator interposed is Ac SA —_— e a(L/2)? 4 (H + LCO)2 We see therefore that the collimator transmits only one-third of the pho- tons directed toward its exit (detector side) hole. We must now account for the fraction of those photons that are produced in the source and directed toward the collimator that actually pass through the source without being degraded. Consider the cross-sectional slice (Fig. D.2) through the cone of thickness dx and at distance x from the 3 sec-l), cone's front face. Denoting the source strength as SV (photons cm™ we obtain for the fraction dF of photons from the slice transmitted dis- tance x dF = 5, Ag eH T g , where AS is the area of the circular source surface facing the detector, fl.is mass attenuation coefficient of Hastelloy N (taken as iron), and D is average density of the MSRE tube bundle. Integrating over the diameter T of the tube bundle we have 203 5,4, L —e™ D2 F o= . (15) WD We can now write an expression for the count rate in the photopeak for those photons that are emitted by the subtended cone (cylinder in this approximation) and transmitted by the collimator. A e« S5 A _ =u DT CR = ___Q___l/__fi__ e 1/3 - (_l_..:e?___l . []_ — eXP(—uGe DG Y] P! P, 4 (H + Lco)z u D © (16) where P' is the fraction of photons transmitted through the heat exchanger shell; other terms have previously been defined. Once we have derived values for the volumetric source strength SV and the average density of the MSRE tube bundle, we can use Eq. (16) to compute the counting efficiency for the heat exchanger. Because our experimental efficiencies were expressed as counts per photon per square centimeter of tube, we assume that each square centimeter of tube in the tube bundle emits 1 photon cm™2 sec”!. From the number of tubes in the bundle and the bundle diameter, we obtain a value of 1.09 cm? of tubing per cubic centimeter of bundle. Thus SV = 1.09 photons cm~ 3 sec™!. Similarly we compute, for the average density of metal in the bundle,-B = 0.976 g/cm3. The count- ing efficiency is where § = 1 photon cm™? sec”!, A computer program was written to evaluate the counting efficiency at several energies over the energy range 100 to 2700 keV and also to compute the counting efficiency for an area source corresponding to the inner sur- face of the heat exchanger shell. The results of our computations are sum- marized in Table D.2, along with the experimentally measured efficiencies. 20U As can be seen, the agreement between calculated and measured efficiency values is excellent between 400 and 1500 keV. The contribution of the heat exchanger shell varies from about 12% at 200 keV to about 4% at 2700 keV., The effect of the shell was not accounted for in our empirical cali- bration but can now be seen to be a negligible factor. Table D.2. Calculated and measured counting efficiencies for the MSRE heat exchanger Energy (keV) Measured Calculated tube bundle Calculated shell D1 D2 100 0.4 0.3 0.08 25.0 26.7 200 1.2 3.3 0.39 -175.0 11.8 300 2.7 3.8 0.36 -40.7 9.5 400 3.9 3.8 0.31 2.6 8.1 500 3.9 3.6 0.27 7.7 7.5 600 3.4 3.5 0.24 =29 6.8 700 3.1 3.4 0.22 -9.6 6.5 800 3.0 3.2 0.20 -6.6 6.2 900 3.0 3.1 0.18 -3.3 5.8 1100 3.0 2.9 0.16 3.3 5.5 1300 2.9 2.8 0.14 3.4 5.0 1500 2.8 2.6 0.13 7.1 5.0 2100 2.7 2.1 0.09 22,2 4.3 2700 2.5 1.6 0.07 36.0 4.3 9p1 = 100 (meas. - calc.)/meas. bDZ = 100 calc. shell/calc. tube bundle. ek SO0~y P W N NN HERFER R NHEOW~OU ™ WN R 23, 24, 25-34, 35. 36. 37. 38. 39. 40. 41. 42, 43, 44, 45, 46-48. 49, 50. 51. 52, 53. 54. 55. 56. 57. un co SO E m:zig nwrHON e * > GGy O;?:fifflgflfqrdpd?>2:C>>»h DU G Sy =S 205 INTERNAL DISTRIBUTION G. Alexander 59, L. Anderson 60. F. Baes 61. C. Bate 62, E. Beall 63. J. Bell 64, Bender 65, S. Bettis 66, F. Blankenship 67. Blumberg 68. G. Bohlmann 69. E. Boyd 70. B. Briggs 71, Cantor 72. L. Carter 73. L. Compere 74, H. Cook 75, B. Cottrell 76. L. Crowley 77. L. Culler 78. J. De Nordwall 79. R. DiStefano 80. J. Ditto 81-83. S. Dworkin 84. F. Dyer 85. P. Eatherly 86. S. Eldridge 87. F. Emery 88. R. Engel 89. E. Ferguson 90. M., Ferris 91. P. Fraas 92, H. Gabbard 93. R. Grimes 94, G. Grindell 95. H. Guymon 96. N. Haubenreich 97. R. Hightower 98. C. Hise 99, W. Hoffman 100. R. Kasten 101. J. Kedl 102, R. Kennedy 103. J. Keyes 104. S. Kirslis 105-106. W. Koger 107, I. Krakoviak 108-110. 111. ORNL-TM-3151 T, S. Kress Kermit Laughon, AEC-0OSR R. B. Lindauer M. I. Lundin R. N. Lyon W. 5. Lyon H. G. MacPherson R, E. MacPherson C. L. Matthews, AEC-0OSR H. E. McCoy H. C. McCurdy H. A. McLain L. E. McNeese J. R. McWherter A, 5. Meyer A, J. Miller R. L. Moore E. L. Nicholson L. C. Oakes A, M, Perry D. M. Richardson R. C. Robertson M. W. Rosenthal H. M. Roth, AEC-ORO Dunlap Scott J. H. Shaffer Myrtleen Shelton M. J. Skinner Din Sood I. Spiewak D, A. Sundberg J. R. Tallackson R. E. Thoma D. B. Trauger H. 0. Weeren A, M. Weinberg J. R. Weir J. C. White G, D. Whitman R. P. Wichner D. Wilson L. V. Wilson E. I. Wyatt F. C. Zapp Central Research Library Y~12 Document Reference Section Laboratory Records Department Laboratory Records, RC 112, 113, 114, 115. 116. 117. 118, llg. 120, 121. 122, 123-132. 133. 134. 135, 136. 137. 138. 139. 140, 141. 142-144. 145-146. 147-163. 164. 165~166. 206 EXTERNAL DISTRIBUTION D. F, Cope, AEC-0SR, Oak Ridge, Tenn., 37830 D. R, DeBoisblanc, Ebasco Services, Inc., 2 Rector Street, New York, N.Y. 10006 C. B. Deering, Black & Veatch, P.0. Box 8405, Kansas City, Mo, 64114 A. R. DeGrazia, AEC, Washington D.C. 20545 N. D. Dudey, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, I11l, 60439 David Elias, AEC, Washington, D.C. 20545 T. A. Flynn, Ebasco Services, Inc., 2 Rector Street, New York, N.Y. 10006 J. E. Fox, AEC, Washington, D.C. 20545 R. Gunnink, LRL, University of California, End East Avenue, P.0. Box 808, Livermore, Calif. 94550 Norton Haberman, AEC, Washington, D.C. 20545 P. A, Halpine, AEC, Washington, D.C. 20545 A, Houtzeel, TNO, 176 Second Ave., Waltham, Mass. 02154 Prof. ir. D.G.H. Latzko, Technological University Delft, Rotterdamseweg 139a, Delft, The Netherlands Fred Marsh, Babcock and Wilcox, P.0. Box 1260, Lynchburg, Va. 24505 J. Neff, AEC, Washington, D.C. 20545 M. Shaw, AEC, Washington, D.C. 20545 R. C. Steffy, Jr., TVA, 303 Power Building, Chattanooga, Tenn. 37401 G. F. Taylor, Fuels and Materials Division, Applied Materials Research Branch, Chalk River Nuclear Laboratories, Chalk River, Ontario, Canada F. N. Watson, AEC, Washington, D.C. 20545 Prof. J. J. Went, Director, N. V. tot Keuring van Electrotechnische Materialen Utrechtseweg 310, Arnhem, The Netherlands M. J. Whitman, AEC, Washington, D.C. 20545 Director, Division of Reactor Licensing, Washington, D.C. 20545 Director, Division of Reactor Standards, Washington, D.C., 20545 Manager, Technical Information Center, AEC Research and Technical Support Division, AEC, ORO Technical Information Center, AEC