LOCKHEED MARTIN ENERGY RESEARCH LIBRARIES | (2 llf I g Ef‘. 3 4456 O45017Y & s i Printed in the United States of America. Available from National Technical Information Service U.S. Department of Commerce 5285 Port Royal Road, Soringfield, Virginia 22161 Price: Printed Copy $5.50; Microfiche $2.2 This report was prepared as an account of work sponsored by the United States Government. Neither the United States nor the Energy Research and Development Administration, nor any of their employses, nor any of their contractars, subcontractors, or their employees, makes 2ny warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. ORNL-TM~4370 UC~-76 — Molten Salt Reactor Technology Contract No. W-7405-eng-26 CHEMICAL TECHNOLOGY DIVISION ENGINEERING DEVELOPMENT STUDIES FOR MOLTEN-SALT BREEDER REACTOR PROCESSING NO. 20 Compiled by: J. R. Hightower, Jr. Other Contributors: C. H. Brown, Jr. E. M. Counce R. B. Lindauer H. €. Savage JANUARY 1976 OAK RIDGE NATTONAL TARORATORY Oak Ridge, Tennessee 37830 operated by UNION CARBIDE CORPORATION for the ENERGY RESFARCH AND DEVELOPMENT ADMINLSTRATION NERGY RESEARCH UBRARIES A 3 uy5k Qus0L7H b ii Reports previously issued in this series are as follows: ORNL-TM-3053 ORNL-TM-3137 ORNL-TM-3138 ORNL-TM~-3139 ORNL~TM~3140 ORNL-TM~-3141 ORNL-TM-3257 ORNL~TM~-3258 ORNL-TM~-3259 ORNL-TM-3352 ORNL~TM-4698 ORNL-TM-4863 Period Period Period Period Period Period Period Period Period Period Period Period ending ending ending ending ending ending ending ending ending ending December 1968 March 1969 June 1969 September 1969 1969 March 1970 June 1970 September 1970 1970 March 1971 December Decembher January through June 1974 July through September 1974 CONTENTS - S’[I}B}{LARLES - . . . - - [% - - - - b . . . . » . - L] - . - . . . . . . . v 1. INTRODUCTION © & ¢ & v o v o 4 o 4 o o & o s o o o« o s o o & » 1 2. CONTINUOUS FLUGRINATOR DEVELOPMENT: AUTORESISTANCE HEATING TEST A.—H:T"'B . " e - . T e s . . . e . . . . . . » . . . . . . . . . L 1 2.1 Experimental Equipment and Procedure, . . « « « « « « + o . 2 2.2 Experimental ResultsS. . + o « + 4 & o v o o o o o o o« o+ 4 2-2-1 I{un AHT—3“1_0 - . . . . . . . . - - . . . . - . . . » 4 2.2.2 Run AHT-3-11 . . & & v v v v e ¢ 4 o o o o + o & o s 4 2.2.3 Run AHT-3~12 . & ¢ v ¢ o v 4 & 4 4 o o o o s o o o 5 2 o 2 2 4 RUH AI'I'I‘"B“.]_B . Py . - - . » . . . - - a - » . . . . . 5 2 ,2-5 RUH AHT‘“B”IA » . . . . . . . . - . . » - . . . . . . 5 2.206 Rlll'l AHT"“B’lS » a . - » - . - . . . . . . » . « - . - 6 2 . 2 - 7 RUII AHT*3"16 » . . ° » . . . . . . . . . . . - . . . 6 2 - 2 » 8 RUII AHT_3M17 . . . . . . . . . . . . . . . . . . . . 6 2.2.9 Run AHT=3-18 . © ¢ ¢ v« 4 v o 4 4 o o o« o s o o o 7 2.2.10 Run AUHT=-3=19. . . o 4 & v v 4 v o o 4 o s o o s 4 7 2-2.11 R'U.II AHT‘"B”ZO. - . . - . . . . . . s . . . . . . . 7 2-2-12 RUI‘l AHT'—'B—Z].- . . . . . - . . » . . . . . . a . . 8 2.3 Discussion, + v v o 4 4« e 4 e e 4 e e e e e e e e e e 8 3. DEVELOPMENT OF THE METAL TRANSFER PROCESS. . ¢ « + ¢ o v o« o + o 9 3.1 CExamination of MTE-3 Equipment and Materials. . . « . . . . 9 3.2 Status of Metal Transfer Experiment MTE-3B. . . + « « + + o 12 L, SALT-METAL CONTACTOR DEVELOPMENT: EXPERTMENTS WITH A MECHANICALLY AGITATED, NONDISPERSING CONTACTOR IN THE SALT~BISMUTH FLOWTHROUGH FACILITY . . *» . . . . ® . = . L4 . . . - . . . . - . - . - » . 14 4.1 Preparation for Mass Transfer Experiment TSMC~7 . . . . . . 14 4.,1.1 Addition of beryliium to the system. « « « « « +» « .« 14 4,1.2 Prerun equilibration of salt and bismuth . . . . . . 14 4.2 Mass Transfer Experiment TSMC-4 . . ¢ « . « v v & « o « o« . 15 4.3 Experimental Results. ¢ « & ¢« ¢ & v ¢ 4 e o 4 « 4« « « « + « 15 - 5. SALT-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A MECHANICALLY AGITATED, NONDISPERSING CONTACTOR USING WATER AND MERCURY. . . . 17 9.1 THhEeOTY. v « o 4 o o o & o o « 2 s & o 5 o 4 o o o 2 « « « . .19 5.2 Experimental Results. « . . « . . « ¢ ¢ ¢ o v o o o 4 v « o« 23 Cn iv CONTENTS (Continued) FUEL, RECONSTITUTION DEVELOPMENT: INSTALLATION OF EQUIPMENT FOR A FUEL RECONSTITUTION ENGINEERING EXPERIMENT. 6.1 Equipment Documentation. 6.2 UY¥_ Generation 6 6.3 Instrumentation and Controls . . . . . . . . REFERENCES. Page SUMMARTES CONTINUOUS FLUORINATOR DEVELOPMENT: AUTORESISTANCE HEATING TEST AHT-3 Twelve additional autoresistance heating runs were made with the AQT-3 equipment. Alr-watey cooling coils were installed on the test sections of the test vessel to eliminate the need to remove insulation during each run. Seven runs were made using both vertical and side~arm electrode test sections, but steady temperature and resistance condifions were not maintained for any appreciable time. Five runs using an electrode at the top of the vertical test section and with the side-arm test section in a frozen position were more successful; temperature control was better, and it was found that steady conditions could be maintained at a much lower salt resistance than previously believed possible. Apparently, the resis- tivity of the salt being used was lower than literaturs values {foadicated by a factor of 9 to 10, probably because of impurities. : DEVELOPMENT OF THE METAL TRANSFER PROCESS We have completed the installation of all equipment for the wmetal transfer experiment MTE-3B in which we will continue to study the steps in the metal transfer process for removing rare-earth fission products from breeder reactor fuel salt. Necessary preoperational checkout of the system is under way before the salts and bismuth will be charged. During this report period, additional sawmples of the salt and bismuth phases (at the three interfaces) from previously operated experiment MTE-3 were analyzed by X-ray diffraction to identify any interfacial impurities. A high concentration of Th02 was found in the LiCl at the Bi~Li interface in the stripper vessel. No oxides were detected at the LiCl--Bi-Th and fluoride salt--Bi~Th interfaces in the contactor. vi SALT~-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A MECHANTCALLY AGITATED, NONDISPERSING CONTACTOR IN THE SALT-BISMUTH FLOWTHROUGH FACTLTTY The seventh tracer run, TSMC~7, has been completed in the mild steel contactor installed in the salt-bismuth flowthrough facility in Building 3592. Prior to the run, approximately 1.5 g~equiv of beryllium was added electrolytically to the salt phase to establish a uranium dis- tribution coefficient of ~ 100. The salt aod bismuth phases were passed through the contactor to ensure that chemical equilibrium was achieved between the salt and bismuth. 237 Mass transfer experiment TSMC-7 was performed after 1 mg of 3 U308 and 11 mg of Mg0 were added to the salt in the salt feed tank. Salt and bismuth flow rates were 152 and 170 cc/min, respectively, with an agitator speed of 68 rpm. Results from the flowing stream samples taken during the run indicate that the salt-phase mass transfer coefficient was 0.0057 + 0.0012 cm/sec. This corresponds to 657 of the wvalue predicted by the Lewis correlation. SALT-METAL CONTACTOR DEVELOPMENT: EXPERTMENTS WITH A MECHANTCALLY AGITATED, NONDLSPERSTING CONTACTOR USING WATER AND MERCURY Data from a series of five experiments performed in the water-mercury contactor have been reanalyzed in an attempt to determine if the apparent change in mass transfer coefficient during the execution of a run was due to the controlling resistance to mass transfer changing from one phase to the other. A model was developed which assumed the reaction under consideration, pp >t [1,0] + za[tg] > 7t [4,0] + PblHg] , to be instantaneous, irreversible, and occurring entirely at the water-— mercury interface. The possibility that the control of mass transfer switched from one phase to the other during a run was also considered in developing the model. vii The model described above was applied to the data obtained from five experiments. Several inconsistencies were found between the model and the experimental data. We concluded that this model does not adequately represent the system, and that further work is necessary in this area. FUEL RECONSTITUTION DEVELOPMENT: INSTALLATION OF EQUIPMENT FOR A FUEL RECONSTITUTION ENGINEERING EXPERIMENT Equipment is described for absorbing UF6 gas into a flowing salt stream containing 'UF4 and reducing the resultant UF5 to UF/ by hydrogena- } tion. Tnstallation of the equipment is under way. 1. INTRODUCTION A molten-salt breeder reactor (MSBR) will be fueled with a molten fluoride mixture that will circulate through the blanket and core regions of the reactor and through the primary heat exchangers. We are developing processing methods for use in a close-coupled facility for vemoving fis- sion products, corrosion products, and fissile materials from the molten fluoride mixture. Several operations associated with MSBR processing are under study. The remaining parts of this report discuss: (1) experiments conducted in a simulated continuous fluorina- tor for studying autoresistance heating in molten salt and formation of frozen salt films on the fluorinator walls; (2) results of dnspection of equipment used in metal transfer experiment MTE-3 for demonstrating the metal transfer process for removal of rare earths from MSBR fuel carvier salt; . 2] (3) measurements of mass transfer of 37U and 972r from MSBR fuel carrier salt to molten bismuth in a mechanically agitated contactor; (4) measurements of mass transfer of lead and zinc between aqueous solutions and mercury amalgams using a mechanically agitated contactor; and (5) design of experimental equipment to be used in engineering studies of fuel reconstitution.’ This work was performed in the Chemical Technology Division during the period September through December 1974. 2. CONTINUOUS FLUORINATOR DEVELOPMENT: AUTORESISTANCE HEATING TEST AHT-3 R. B. Lindauer A drain line was installed from the side arm to the wvertical section of the test vessel in order to remove the salt heel from the side arm. This was done to facilitate removal of the electrode section which had melted off during autoresistance heating test run AHT-3-9; it was then possible to resume autoresistance heating test No. 3 (AHT-3). Air-water cooling coils were installed on the side arm and vertical test sections of the test vessel to eliminate the need to remove (or loosen) insulation during each run. The cooling coils were placed on the test section in a coaxial position because of the arrangement of existing heaters. Temperature control could be improved by using heating and cooling zones which are transverse to the axis ol the test section. Seven runs were attempted using the entire test vessel (AHT-3-10 to ~16). The best operation was attained in run AHT-3-15, but the resistance decreased after about an hour of autoresistance heating. The maximum power reached was 1130 W. The only temperature point above 350°C at this time was at the junction of the electrode side arm and the vertical test section. An electrode was installed at the top of the vertical test section, and five more ruas were made with the side arm frozen. Satisfactory operation was achieved, but with an unusually low salt resistance (0.09 to 0.12 ). This was only about 10% of the resistance found during good operation with LizBeF4. All attempts to transfer salt from the test ves- sel, which had frozen salt on the wall, were unsuccessful. A different autoresistance power supply with a higher current capacity was used in the last two runs, since large currents (100 A) were being used at these low resistances. It was found that steady conditions could be maintained with a resistance of 0.095 @ using 110 A. It is bhelieved that both the low resistance and the salt transfer difficulty are associated with crud and impurities in the salt. 2.1 Experimental Equipment and Procedure The equipment and flowsheet were essentially the same as described previously.l A 1/2~in.-0D nickel drain line was installed from the low point of the electrode side arm to the side of the 6-~in. vertical section of the test vessel, 9 in. above the botltow. The line was installed to permit draining the salt heel from the side arm so that the section of electrode which melted off in the last experimentl could be recovered. During normal operation this line will not be heated so that a salt plug will form. A new electrode was fabricated consisting of a 3/4-in. sched 40 nickel pipe with a 5-in.-long section which extended up the center of the glanting portion of the electrode side arm. This electrode is there- fore similar to the one which will be used in the next experiment, AHT-4, in which the circulating salt will enter the test vessel through the electrode. Another change made to the equipment was the installation of cooling coils on the electrode side arm and on the vertical test section. These coils were strapped on the 6-in. pipe between the Calrod heating elements. There were four separate cooling circuits. Two coils {top and bottom) were installed on the slanting section of the side arm, connected in parallel, and had the same air-water supply. Four coils (north, east, south, and west) were installed on the vertical test section. These were connected with three separate cooling supplies. A very low (less than 50 cmB/min) water flow was metered into the cooling air to increase the heat removal capacity. Exit temperatures of each coil were monitored, and the water flow rate was adjusted to the maximum which could be vaporized. This was done by maintaining the exit cooling temperature slightly above 100°C. This cooling system eliminated the need to remove insulation from the test section in order to form the frozen salt film. The cooling rate was also increased. The vent line in the center of the top flange of the 8-in.-diam disengaging section was moved to allow this access nozzle to be used for both an insulated thermocouple during runs AHT-3-10 to -16, and for the autoresistance electrode during runs AHT-3-17 to -21. 1In the last five runs, a 1/4-in.-diam nickel rod was used as the electrode with the rod submerged 4 in. into the salt. The length of the test section was approxi- mately 24 in., compared with 44 in. for the side-arm electrode. These lengths assume a current path to the wall just below the gas inlet side arme. Operation of the system was similar to that previously described. The entire test vessel was heated to between 500° and 550°C. Heat was then turned off the test section, and cooling air and water were turned on. When the wall temperature of the test section reached 350°C (about 2 hr), the autoresistance power was inctreased stepwise until the electrical resis- tance between the electrode and the molten salt at the bottom of the test vessel reached a constant value. The autoresistance power was turned on at a very low power (less than 10 W) at the start of cooling so that the resistance could be calculated continuously from the voltage and current. An instrument was installed to perform this division electrically, and the quotient was recorded continuously on a 0 te 10 mV recorder. When tne resistance had lecveled off, the cooling was adjusted by adjusting the water and air flows. The voltage to the autoresistance circuit also required adjusting to maintain a constant frozen film. 2.2 Experimental Results 2.2.1 Run AHT-3-10 test section. Insulation was loosened to cool the vertical test scctiom. The electrical resistance through the salt was calculated periodically from the voltage and current. Sufficient water was added to the cooling aly to keep the exit air temperature near 100°C. When the test section temperatures reached 350°C the autoresistance power was left on. The resistance remained fairly constant at about 1.7 § for 30 minn as the power was increased to 2.1 kW (60 V and 35 A). At this point, the current rose sharply and the resistance dropped. The temperature at the top of the electrode side arm vose 20°, indicaring an electrical short at this point. After the hot spot was cooled for 15 min, autoresistance heating was resumed. The resistance slowly decreased however, and sufficient power could not be introduced to keep the vertical test section from freezing. After run AHT-3-12, it was discovered that one of the temperature recorders was reading 50°C too high, and the run had been started with too low a temperature at the bottom of the test vessel. 2.2.2 Run AHT-3-11 P e S S Conling coils were installed on the vertical test section before this run. The system was heated to a higher temperature than normal to check out the new coolers. The average cooling rate was increased from 55 to 85°/hr, but there was time for only 20 min of autoresistaoce heating, not enough to reach steady conditions. The resistance dropped from 0.2 to 0.1 @ during this period. Again, because of the incorrect temperature recorder, the bottom of the test vessel was allowed to drop below the liquidus temperature (505°C). 2.2.3 Run AHT-3-12 Cooling was started with the test vessel aft a higher temperature than normal--the test sections were about 600°C. A good cooling rate of about 98°C/hr was achieved, and the autoresistance power was started early in the day. Current was increased stepwise from 20 to 30 to 40 A, The resistance started to drop at 40 A, although a maximum power of only 480 W was being used. The resistance was 0.25 §i. Since the resistivity of the salt was believed Lo be 0.75 {-cm, this indicated a very thin film. The power was reduced in an attempt to vaise the resistance, but the test section froze over. 2.2.4 RBun AHT=-3-13 Before this run was started, it was discovered that one temperature recorder was reading 50°C higher than the other. When this was corrected it was found that the electrode side-arm test section was 50°C hotter than the wvertical test section. There was insufficient time to balance the temperatures; by the time the side arm was cooled to 350°C, the verti- cal test section had cooled to 280°C. Autoresistance power was turned on, although the resistance was still low, and shorting occurred in the side arm. 2.2.5 Run AHT-3-14 Cooling of the electrode side arm was begun an hour before the cooling was started on the vertical test section in an effort to form a frozen film on the side arm before the vertical section became too cool; however, the vertical section cooled at a much higher rate. The junction 0of the vertical and side-arm test sections cooled very slowly. By the time this point reached 350°, the vertical test section had frozen com- pletely. A recording resistance meter was installed before this run to provide an immediate indication of resistance changes. 2.2.6 Run AHT-3-15 Autoresistance heating was started when the electrode side-arm test section reached 350°C, although the vertical test section was still over 400°C. The resistance was unusually high, about 2 ¢, although the junction of the vertical and side—arm sections was about 440°C. The power was increased stepwlise with only slight changes in resistance until 1130 W (38 V) was reached, at which point the resistance suddenly dropped to 0.7 §. Autoresistance heating of greater than 100 W had been applied for about 70 min, but temperatures on the test-section walls were still decreasing rapidly. 2.2.7 Bun AH_I_’::‘_S:} é An attempt was made to duplicate run AHT-3-15, but with the junction of the vertical and side~arm test sections below 350°C. Autoresistance heating was not applied until this temperature was reached. When voltages approaching 20 V were applied, the resistance dropped from v 2 to ~ 1 Q. It was still believed that the resistivity of the salt was about 0.75 O- cm; the resistance was not allowed to get below 0.8 @, corresponding to a 3/4~in. film. TIf, at this time, it had been known that the resistivity of the salt was much lower, satisfactory operation might have been achieved with the electrode side arm by applying sufficient power to balance the heat loss before temperatures became too low and the test section froze over., 2.2.8 Run AHT-3-17 The side arm was allowed to freeze completely, and autoresistance heating was applied from an insulated 1/4-in.-diam nickel rod janserted through the 8-in.~diam disengagement section above the vertical test scc~ tion. The rod extended 4 in. into the salt. The resistance again increased to 800 W. This was similar to previous runs with the electrode side arm. However, during this run, as the resistance dropped, heating was maintained and relatively stable operation was achieved at a power level of about 900 W. The resistance decreased slowly, reaching 0.11 @ at the end of the operating periocd. The current capacity of the auto- resistance power supply (100 A) was nearly reached. Test-section wall temperatures had leveled off at an average temperature of about 275°C. 2.2.9 Run AHT-3-18 In this run, we not only attempted to duplicate conditions of the previous run, but also to reach steady conditions early enough in the day to transfer the molten-salt core in the test section to the feed tank in order to visually check on the salt~film thickness. Operation was very similar to run AHT-3~17, with a resistance of 0.12 2 using 850 W of autoresistance power. We were unable to tramnsfer the salt as planned, although ten pressure-vacuum cycles were applied to the feed tank. The test vessel bottom temperature and transfer line temperature indicated a slight salt movement, and the applied pressure showed a slight indication on the test vessel level instrument. The salt level did not fall below the agutoresistance electrode. 2.2.10 Run AHT-3-19 The two previous runs were more or less duplicated in this run. The test vessel bottom was heated to a higher temperature at the end of the run to aid in transfer, but this was not successful. As in the two pre- vious runs, the current capacity of the autoresistance heating suapply was not sufficient, and wall temperatures continued to decrease at the end of the run. 2,.2.11 Run AHT-3-20 The autoresistance power supply was increased from 100- to 200~A max current; however, the new supply had only a 24-V max instead of 100 V. During this run, the resistance was allowed to increase to 0.8 before autoresistance heating was applied, and only 600 W of power was available. This was not sufficient to prevent the molten core from freezing, as was indicated by the test section liquid-level instrument going off the scale and an increase in the resistance to > 3 . 2.2.12 Run AHT-3-21 The test-section wall was cooled as rapidly as possible and auto- resistance heating was started when the salt resistance was 0.37 2. All wall temperatures were below 350°C at this time. A maximum resistance of 0.40 @ was reached when the maximum 24 V was applied to the electrode. The vesistance then decreased and the power rose to a maximum of 2500 W. The voltage was decreased, and stable operation was reached with a resis-— tance of about 0,09 & and 1470 W. The wall rfemperatures also leveled out. The resistance decreased very slowly and voltage and power were conse- quently lowered to 1160 W. The temperature of the bottom of the test vessel was being raised at this time to facilitate transfer of the salt from the test vessel for inspection of the frozen filwm; this was probably the cause of the decrease in resistance. After the power was decreased, the resistance slowly increased to 0.095 Q. We were still unable to transfer salt after repeated attempts, so heating of the entire test vessel was begun to remove all of the salt to the feed tank. During the heatup, inspection of the cell revealed a salt leak which was located at the junction of the salt transfer line and the bottom of the test vessel. Approximately 6 kg of salt was lost. 2.3 Discussion Successful operation might have been achieved in most of these runs if it had been known that the resistivity of the salt was not 0.75 Q-cm, as stated in the literature,z but was considerably lower. In many cases, the autoresistance power was reduced when the resistance was in the range of 0.5 to 1.0 &; this was done to prevent melting the frozen film. Later operation without the side arm showed that steady operation could be majintained in the 0.09- to 0.15-0 range, and that the resistivity of the salt was probably lower than 0.75 Q-cm by a factor of 5 to 10. Operation with the electrode in the side arm was unsuccessiul for two reasons. Cooling of both vertical and slanting test sections at the same rate to avoid freezing one section before a complete film was formed on the other section was quite difficult with the installed cooling zones. In addition, the junction of the side arm and vertical section was diffi- cult to cool, and shortiog occurred at this point until an air jet was applied. Both of these difficulties should be lessened considerably by circulating salt through the test vessel. This would be expected to pro~ mote a more uniform temperature over the entire test section. Conse- quently, we have started designifig a system which will allow salt to be circulated through the test vessel. 3. DEVELOPMENT OF THE METAL TRANSFER PROCESS H. C. Savage Engineering experiments to study the steps in the metal transfer process for removing rare-earth fission products from molten-salt breeder reactoy fuel salt will be continued in new process vessels which dupli- cate those used in a previous experiment, MTE~3.3 Installation of the equipment for the new experiment, degignated MTE~2B, was completed during this report period. Experiment MTE-3B utilizes mechanically agitated contactorsé to achieve effective mass~transfer rates of the rare-earth fission products between the salt and metal phases in the metal transfer process {(as was done in ezperiment MTE-3). 3.] Examination of MTE-3 Equipment and Materials We have previously reported the results of analyses of the salt and bismuth phases from experiment MTE*-ZLb These analyses were made in an attempt to determine whether or not the Jlower-than~expected mass transfer coefficients observed in experiment MTEwBB were due Lo the presence of films (composed of solids) at the salt-bismuth interfaces formed by entry of impurities, such as oxides, into the system. These analyses, which included chemical analyses, metallographic examivation, electron beam scanning, and X-ray fluorescence avalyses, indicated the presence of & 3 10 significant concentrations of iron (up to 3500 ppm) and thorium at the interfaces. Visual examination of the 2-in.-diam plugs removed from the three interfaces (fluoride salt—--Bi~Th, LiCl--Bi-Th, and LiCl--Bi-Li) indicated the presence of a layer of material ~ 1/32-in, thick at the interface between the LiCl and Bi-Li in the stripper vessel. No foreign material was seen at the interfacial areas between the fluoride salt—--Bi-Th and LiCt1-~Bi-Th in the contactor. During this report period, samples of material were removed from the vicinity (within 1/32 in.) of each of the three salt-metal interfaces in the MTE-3 equipment and analyzed by X-ray diffraction. The results of the six analyses are given in Table 1. The only oxides in the system were at the interface between LiCl and the Li-Bi stripper alloy in the stripper vessel. A high concentration of ThO. was found in the TiCl at 2 this interface. Although no thorium should have been present in the stripper during most of the operating time, thorium could be expected to combine with any oxides present when the fluoride salt (containing thorium) was entrained into the chloride salt. No oxides were detected at the LiCl--Bi-Th interface in the contactor. The analysis of the mass transfer rates observed during this experiment suggested that some hindrance of mass transfer was occurring at this interface, possibly caused by oxide films. Analyses of the material near this interface, however, do not support the theory ihat oxide films at the LiCl--Bi-Th interface slowed down the mass transfer. The samples taken for X-ray diffraction analysis were extremely small (about 2 to 3 mm3 of material was removed from the interface for each sample), and effects of segregation during freezing will make interpre- tation of the results of these analyses very difficult. For instance, it is interesting to note that no LiCl was detected in the sample of LiCl taken from the interface in the stripper vessel. Bismuth was detected in each of the salt samples analyzed by X-ray diffraction. 1t is likely that during freezing, bismuth was forced into the already frozen salt and was not present in the salt when both phases were molten. 11 Table 1. Results of X-ray diffraction analyses of material removed from the vicinity of salt-metal interfaces in experiment MTE-3 A st = e e Identified Estimated composition Phase material (mole %)% 1iCl in stripper Bi 25-50 L13ihf7 10-30 Th02 40-80 unidentified solid - solution Li-Bi alloy in stripper Bi 75-95 BeF2 2-10 Bi—-Th in contact with Bi 90 LiCl in contactor LiCl in contactor Bi 30-70 LiCl 30-70 LlBth7 10-30 Fluoride salt in con- Bi 20-40 tactor Li.ThF 40-80 3 7 L17Th6}31 5-15 Bi~-Th in contact with Bi. 60-90 fluoride salt in con- Th (unidentified L tactor compound) Fe (no X~ray data trace available) a . . - X-ray diffraction can only detect > %~ 5 mole Z. It is to be noted that iron and/or iron oxides were not found by X- ray diffraction, since the amounts present (based on chemical analyses of similar samples) are below the limit of detectability by X-ray diffraction (v 5 mole %). An additional chemical analysis was made of the 1/32-in.-thick layer of material found between the LiCl and the Bi~Li in the stripper vessel. This material had a different stiucture and a dark gray or black appear- ance. The followiag resulis were veported (wt 7%): Bi - 51.24 ' — unable to analyze Th ~ 3.7 Fe - 14.0 Li - 5.0 O, - unable to analyze due to high B8i Bi - 0.23 Cl ~ 394 ppm The very high ircn concentration (14 wt %) is unexpected. Previous anal- yses of other samples of similar appearing material indicated irou concen- trations of about 0.25 wt 7. It is not possible at this time to draw firm conclusions about the relationship of these observations to the low mass trvansfer rates seen in experiment MTE~3. The transfer of fluovide salt into the chloride salt just prior to shutdown, and the length of time between shutdown and inspection (from February 1973 tu February 1974), introduce much uncer- i~ L tainty in an accurate interpretation of the analyses. We plan to closely J monitor the salt and metal phases duvrinyg operaiion of exveriment MTE~ L L o B < for buildup of impuritics (iron and oxides). Sampling phases at or uear cach interface will be atrempted. 3.2 Starus of Metal Transier bxperiment MTE—-3B Installation of the new process vessels and squipment for metal trans- fer experiment MTE-3B was completed during this report period. Figure 1 is a photograph of the process vessels afier electric heating elements and thermocouples were installed. We have completed the process piping installation, thermocouple and heater hookup, and the calibration of the temperature recorders and controllers, pressure gages, selecited thermo- couples, and all gas and cooling-water rotameters, The experiment is now ready for preoperational checkout, leak testing of the vessels and piping, pressure tests at the operating temperatuve of 650°C, and hydrogen treat- ment of the interlor of the carbon stee! process vessels to remove oxide impurities before charging salts and bisauth. s A S A S e P — Pig. L. ment MTE-3B. L — a8 PHOTO NO. 2706-74A | | . Ko ,SALT—METALi / i CcONTACTOR ° Photograph of processing vessels for metal-transfer experi- 14 4. SALT-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A MECHANICALLY AGITATED, NONDISPERSING CONTACTOR IN THE SALT-BISMUTH FLOWTHROUGH FACILITY C. H. Brown, Jr. We have continued operation of a facility in which mass transfer rates between molten LiF—BeFZ—ThF4 (72-16-12 mole %) and molten bismuth can be measured in a mechanically agitated, nondispersing contactor of the "Lewis" type.6 A total of seven experimental runs have been completed to date. Results from the first six runs have been reported previously. Preparation for and results obtained from the seventh run, TSMC-7, are discussed in the following sections. 4.1 Preparation for Mass Transfer Experiment TSMC-7 Prior to the run, it was necessary to: (1) add beryllium to the salt to adjust the uranium distribution coefficient, and (2) contact the salt and bismuth by passing both phases through the mild steel contactor to ensure that chemical equilibrium was achieved between the salt and bis- muth. 4.1.1 Addition of beryllium to the system As discussed previously,7 it is desirable to maintain the uranium distribution coefficient at a relatively high level (> 30), so that the actual value of the distribution coefficient will not affect the calcu- lation of the overall mass transfer coefficient. Approximately 1.37 g-equiv of beryllium was added electrolytically to the salt phase in the treatment vessel (T5), which raised the uranium distribution ratio to > 97. 4.1.2 Prerun equilibration of salt and bismuth 237 . A U308 tracer was utilized to measure mass transfer rates across the salt—bismuth interface in the stirred interface contactor while the system was otherwise at chemical equilibrium. In order to ensure chemical equilibrium between the salt and bismuth, both phases were passed through 15 the contactor prior to run TSMC-7. Three attempis at phase equilibration runs were necessary before a satisfactory flowthrough was achieved. During the first two attempts, a leak developed in the transfer line from the bismuth feed tank (T1l) toe the contactor. This transfer line was completely replaced, along with the associated Calrods and thermal insu- lation. The third attempt at phase equilibration was successful with salt and bismuth flow rates of 152 cc/min and 170 ce/min, respectively. The agitator was operated at ~ 100 rpm. 4.2 Mass Transfer Experiment TSMC-4 After the phase equilibration described above was accomplished, the salt and bismuth were transferred to their respective feed tanks. 237 An 11-mg quantity of tracer, consisting of 1 mg of U,0, and 10 mg 378 of MgO which had been irradiated in the ORR for 72 hr, was placed in a steel addition vessel. It was then inserted in the salt-feed tank where it was sparged with ~ 0.5 scfm of argon for 1 hr to facilitate dissolu~ tion of the tracer in the salt phase. The volumetric flow rates of salt and bismuth to the contactor were set at 152 cc/min and 170 cc/min, respectively, by controlled pressuri- zation of the feed vessels (T3 and Tl). The stirrer rate was set at 68 rpm for the run. Seven sets of samples were taken of the salt and bis~ muth effluent streams from the contactor. 4.3 Experimental Results The samples taken during the run were analyzed by first counting the 2 - 37U (207.95 keV B ). The material in the sample capsules was then dissolved, and the activity of 237U was sample capsules for the activity of counted again. The counting data obtained in this manner are shown in Table 2. The overall salt-phase mass transfer coefficient was calculated using three different equations derived from an overall mass balance around the 7 contactor. The average measured mass transfer coefficient, with the Table 2. Counting data obtained from run TSMC-7 Solid analysis Solution analysis Solid analiysis Solution analysis Sample for 2370 for 237y Samplie for 237y For 237g code? {counts/qg} {counts/qg) coded (counts/qg) {counts/qg) Samples taken prior to run 313-B-5 2.68 x 107 6.39 x lo* 311-5-5 < 4.2 x 107 < 1.8 x 10Y 314-B--_ 2.11 x 107 6.77 x 1ot 312-5-5 < 5.2 x 107 < 1.5 x 10" 317-B-1 2.25 x 10% 6.79 x 10" 315-5-3 < 6.6 x 10° < i.8 w10 318-B-1 2.57 x 10% 6.47 x 10" 316-3-3 < 7.6 x 107 < 9.5 x 103 Samplies taken prior to run but after addition of tracers 319-5-3 3.95 x 10° 5.46 x 107 320-5-3 3.95 x 10° 5.61 x 105 Samples taken during run 321-B-FS 1.20 x 10° 3.91 x 10° 329-S-F§ 2.21 x 10% 2.89 x 10° 322-3-F8 1.36 x 107 4.33 x 10° 330-5-58 2.85 x 10% 3.32 x 10° 323-B-¥s 1,47 x 10° 4.81 x i0° 331-8-FS 2.75 x 10° 3.26 x 10° 324-B-FS 1.81 x 107 5.03 x 10° 332-5-75 2.89 x 10° 3.97 x 10° 325-B-FS 1.62 x 10° 5.06 x 10° 333-5-7g 2.82 x 108 2.96 x 10° 326-B-FS 1.79 x 10° 5.05 x 10° 334-5-FS 2.93 x 10° 3.90 x 106 327-B-FS 1.78 x 10° 5.17 x 10° sampies taken after run 335-B-1 8.65 x 10* 2.22 x 10° 33¢-5-3 -- -~ 336-B-1 6.66 x L0Ob 2.18 x 10° 340-5-23 -- - 337-B-2 1.24 x 10° 3.78 x 10° 341-5-4 -- -— 338-B-2 1.34 x 107 3.86 x 10° 342-5-4 1.64 x 108 2.19 x 10° 343-B-5 i.99 x 10° 5.05 x 10° 345-5-5 < 9.8 x 107 < 8.7 x 103 344-B-5 1.84 x 107 5.58 x 10° 346-5-5 < 5.3 x 107 < 1.7 x 1% Each sample 1is cesignated by a code corresponding to A-B-C, where A = sample number; B = material in sample (B = bismuth, S = salt): and C = sample origin: 1 =71, 2 =72, 3 ="T3, 4 = T4, 5 = T, PS8 = flowing stream sample. o1 17 corresponding standard deviation, is 0.0057 + 0.0012 cm/sec. This value . —_— . 6 corresponds to 657 of the value predicted by the Lewis correlation. A Lewis plot of the results for this run, along with the results from cuns TSMC~3 through -6, are shown in Fig. 2. The nomenclature used in Fig. 2 is: k = individual phase mass transfer coefficient, cm/sec, . . . . 2 v = kinematic viscosity, ¢m /sec, 2 . . Re = Reynolds number (ND /v), dimensionless, D = stirrer diameter, cm, N = stirrer rate, 1/sec, subscripts 1, 2 = phase being considered. In Fig. 2 it can be seen that the mass fransfer group based on uranium is 45 + 13% of the Lewis correlation for runs 3, 5, and 7, while that same group is 103 + 47 of the Lewis correlation for runs 4 and 6. The mass transfer results based on zirconium are consistently slightly less than the values based on uranium. 7This discrepancy is probably related to the inability to correct for the self absorption of the 743.37 keV B~ from the 97Zr in the s0lid bismuth samples. We believe that at a stirrer speed bétween 160 and 180 rpm, the ini- tiation of entrainment of salt into the bismuth begins to cccur; the apparent increase in mass transfer coefficient is a manifestation of an increase in the surface area for mass transfer due to surface motion. Experiments with water—-mercury and water-methylene bromide systems support this belief. 5. SALT-METAL CONTACTOR DEVELOPMENT: EXPERIMENTS WITH A MECHANICALLY AGITATED, NONDISPERSING CONTACTOR USING WATER AND MERCURY C. H. Brown, Jr. The reference flowsheet for the proposed MSBR processing plant calls for the extraction of rare earths from the fluoride fuel carrier salt to an intermediate bismuth stream. One device being considered for performing this extraction is a mechanically agitated, nondispersing contactor in 18 ORNL-DWG 75-15156 100 - [ T T T T T TTT7 - /g‘rsmcw - - TSMC-6/ & rcric-a . i g TSMC-6 - LEWIS / ®TSMCS CORRELATION— L \7’ @TSMC-3 = / &TSMC-5 / N 10 - / - - / ®TSMC-7 ~ 60K, / ®2°7y /i | i | L1 L 1tif | OO0 10000 10000 Re|+72/% Rep Fig. 2. Comparison of mass transfer coefficients measured in the salt-bismutlhh contactor using 2379 and ?/7r tracers with values predicted by the lLewis correlation. 19 which bismuth and fluoride salt phases are agitated to enhance the mass transfer rate of rare earths across the salt-bismuth interface. Previous reports ~10 have shown that the following reaction in the water-mercury system is suitable for simulating and studying mass transfer rates in systems with high density differences: i 24 P77 [H,0] + zaltg] > zn” [H,0] + Pb[ig] . (1) i 8 . A large amount of data have been reported £for the water-mercury system in which it was assumed that the limiting resistance to mass transfer existed entirely in the mercury phase, as suggested by literature corrve- lations. An experiment was previously describedll that was designed to iden- tify the phase in which the controlling resistance to mass transfer resided. In these experiments, it was noticed that the mass transfer coefficient appeared to vary during the course of a run if the initial concentrations of Zn in the mercury phase and Pb?’+ in the water phase were not equal. The mass transfer coefficient would abruptly change values after the experiment had been in progress for about 12 to 18 min. During this report period, the data from the experiments that had been performed in the water-mercury contactor were reanalyzed in an atfempt to determine if the apparent change in mass transfer coefficient was due to the con- trolling resistance changing from one phase to the other during the course of an experiment. 5.1 Theory . 11 . As was previously reported, the reaction being studied, Eg. (1), is considered to be a fast, irreversible, ionic reaction which occurs entirely at the mercury-water interface. The rate of transport of reactants from the bulk phase to the inter- face where they react is given by: N = 1 e 7n° K (Cne,p — € Zn®,i ) s (2) 20 2+ = N Ky ) A(CPb2+ — C,.2+ ), (3) 0 , B Pb™ ,i where = individual phase mass transfer coefficient, cm/sec, = rate of mass Cransport to the interface, g/sec, 2 interfacial area, cm , = concentracion, = O B 2w | = denotation of bulk phase concentration, i = decotation of iaterfacial conceatration, g/em”, Hg,H20 = refers to the phase being considered. . , . 2+ . Since one mole of 7n° is equivalent to one mole of Pb according to Eq. (1), then NPb2+ = NZn°° Substituting Eqs. (2) and (3) into this expres- sion yields the following equality, ) = k 2+BWC 2+ ) 4) ppt i) (4) 1.0 Cpp k -0 Hg(CZn°,B Zn°,1i ” L . As earlier reported, it the controlling resistance to mass transfer is assumed to be in the ore phase, the interfacial reactant concentration in that phase is very swall with respect to the bulk reactant concentra-— tioa. In the transient experiment which we bave performed,ll it may be possible for the limiting resistance to mass transfer to switch from one phase to the otner during the course of an experiment. Since both inter- facial reactant concentrations are very small with respect to their bulk phase concentrations, the following equality exists at this transition point: k o o, = K 2+ . Hg CZH , B kHZO CPb , B (5) If the controlling resistance to mass transfer is in the aqueous , 2t phase, two equations, one for the concentration of Pb and one Ior the . 2+ . concentration of Zn , can be derived: 'CPb2+(t) kazo A 1n | ===} = — —— t , and Cpp2+(0) 1.0 - . k.. A g LPb2+(O) C2n2+(t> _ MV“EEO - 3 C - - I pp2+(0) | Yi.0 2 where V = volume of phase, cm’, t = time, sec, and parentheses iudicate quantities which are functions of time. (&) (7) [f the mervcury phase contains the controlling resistance to mass - . - . L, 2t transfer, two other equations, one for the concentration of Pb i ., 2 . for the concentration of Zn~ , ¢an be derived: [ C G) —C +{(0) 4+ C +(t k A (Jzno(") Pb2+‘(J) P1£32+( ) ~ He — 1o = w‘“”fi"“'t, and C D(O) V‘, 3 Zn He v CZn°(O) *‘Czn2+(t) } kH A | In c. . (0) Ty t i 7n° Hy Equations (6)-~(9) are all of the form, y = -mt , where y = the natural logarithm of a concentration ratio, = a mass transfer coefficient times the interfacial area divided by the phase volume, and £t = time, and one (8) (9) (10) This indicates that a logarithmic plot of the concentration ratio vs time should yield a straight line, with the slope belng proportional to the mass transfer coefficient. If the control of mass transfer switches fFrom 22 one phase to the other during the execution of an experiment, then the slope of the semilog plot mentioned above would indeed change, as was observed. Equation (5) can be used with experimental values of k and He? kH 0’ o concentrations of an+ and Pb2+, which are evaluated at the time of apparent change of mass transfer coefficient and at the beginning of the experiment, to determine the phase in which the limiting mass fransfer resistance lay at the beginning of the run. The first step in the detor- mination of the mass transfer limiting phase at the beginning of a run is to find the pair of individual phase mass transfer coefficients, one before and one after the transition point, which satisfy Eq. (5), with the bulk reactant concentrations evaluated at the transition point. Two combinations of mass transfer coefficients are possible, giving Eq. (5) the following two forms: (1) . (2) ng C oo = kH 0 CPBZT , and (11) tr 2 tr (2) (L k C. o = k C. . 24 . (12) Hg Z1i e H20 Pb r where superscripts 1 and 2 refer to the slope in Eq. (10) before and after the transition, respectively, and the subscript tr refers to the transi- 2 tion point. Equation (6) or (7) is used to determine k (1) and k(”), 1.0 H,0 (1) and k(2) 2 2 and Eq. (8) or (9) is used to determine ng’ He Once it has been determined which of the two equations, FEq. (11) or (12), is satisfied for a particular run, the mass transfer limiting phase at the beginning of the run may be determined by further application of an inequality similar to Eq. (5). The individual phase mass transfer coef- ficient in the limiting phase times the bulk reactant concentration in that same phase must be less than the same product in the nonlimiting phase. Therefore, if Eq. (11) is satisfied, then the direction of the following inequality determines the initial mass transfer limiting phase, 23 > . (2) . CZn°(O) - ]‘1{20 CPb2+(O) . (13) (1) kHr Similarly, if Eq. (12) is satisfied, then the limiting phase is determined by the direction of the inequality: (2) > (1) \ ng €, ne(0) < kHZO Cop2+(0) (14) 5.2 Experimental Results Five experiments were run in the water-mercury contactor to determine if the apparent change in mass transfer coefficient was due to controlling resistance changing from one phase to the other during the course of an experiment., The initial mercury~phase zinc concentration was held constant at 0.1 M. Phase volumes and agitator speed were also held constant at 1.8 liters and v 150 rpm, respectively. The experimental data obtained from these five runs, the conditions under which the runs were made, and the transient response of the system in terms of aqueous lead and zinc concentrations are shown in Table 3. The data were analyzed by use of the equations in the previous section. Results of this analysis for determination of the phase which con- tained the limiting resistance to mass transfer at the beginning of each run are shown in Table 4, along with the measured value of the mass trans- fer coefficient in the limiting phase. These results indicate that the limiting resistance to mass transfer was initially in the mercury phase in the runs made without initial aqueous-phase lead concentration eaqual to or less than the initial mercury-phase zinc concentration. For the runs made with an initial lead ion concentration greater than the initial zinc amalgam concentration, the aqueous phase controls mass transfer. The results from run 148 (in which the initial reactant concentrations are nearly equal) are difficult to analyze, because only a slight change of slope in a plot of Eq. (10) was detected. Tn runs 145 and 149, the initial lead ion concentration was low with respect to the zinc amalgam concentration. This would indicate that the 24 Table 3. Concentration vs time data from the water-mercury contactor Initial Hg-Zn analgam concentration = 0.1 M Stirrer velocity = 150 rpm Phase volumes = 1,8 liters each Paddle diameter = 7.62 cm Paddle height = 1.91 cm Run Elapsed time Lead ion Zinc ion numbe {min} concentration (M) concenitration (M) 145 0 0.0197 0.0003 3 0.0171 0.0038 6 0.0141 0.0070 9 ¢.0112 0.0095 12 0.0092 0.0115 15 0.0071 0.0134 18 G.0053 0.0158 21 0.0042 0.0170 24 0.0033 0.0181 27 0.0025 0.0186 30 G. 0019 0.0196 146 0 0.141 0.00021 3 0.131 0.0187 6 0.117 0.034¢6 5 0.101 0.0478 12 0.0883 0.0604 15 0.0738 0.0750 i8 0.0589 0.0826 21 0.0589 0.0887 24 0.0550 0.0210 27 0.0526 0.0948 30 ¢.0512 0.0979 147 0 0.184 0.00104 3.33 0.171 0.0338 6 0.144 0.0520 9 0.132 0.0678 12 0.121 0.0811 15 0.109 0.0872 18 0.103 0.0%02 21 0.101 0.0941 24 0.0999 0.0979 27 0.0975 0.0971 30 0.0956 0.100 148 o 0.0960 0.0007 5 0.0749 0.0199 10 0.05h68 0.0375 15 0.0419 0.0516 20 0.0331 0.0631 25 0.0249 0.0704 30 0.0176 0.0784 35 0.0148 0.0834 40.33 0.0102 0.0899 45 0.0670 0.0906 50 0.0046 0.0922 55 0.0029 0.0922 60 0.0017 0.0948 1499 0 0.0192 0.00045 3.17 0.0156 0.00512 6.07 0.0122 0.00795 9.25 0.0970 0.0106 12 0.0741 0.0130 15 0.00536 0.0148 18 0.00384 0.0162 21.08 0.00249 0.0177 24 0.00211 0.0188 27 0.00127 0.0190 30 0.00069 0.0199 25 Table 4. Initial lead ion concentration and mass transfer coefficients in the phase initially containing the limiting resistance to mass transfer for runs 145 to 149 Mass transfer coefficients in the Initial lead phase controlling mass transfer Run ion concentration initially (cm/sec) number (_M;) ng kH? 0 145 0.0197 0.0012 149 0.0192 0.0013 148 0.0960 0.0056 146 0.141 0.0061 147 0.184 0.0046 aqueous phase should initially contain the limiting resistance, since the initial lead dion concentration times the mass transfer coefficient should be less than the initial zinc amalgam concentration times the mass trans-— fer coefficient in that phase. Similarly, in runs 146 and 147, the ini- tial lead ion concentration was large with respect to the zinc amalgam concentration; this would indicate that the mercury phase should initially contain the limiting resistance to mass transfer. According to the model we have developed, if the aqueous phase ini- tially contains the limiting resistance to mass transfer, the following inequality is satisfied: kyy Cgno(0) > kHZO Cpy24(0) (15) During the course of an experimental run, the inequality shown in Eq. (15) should always be satisfied, since the lead ion concentration decreases with an equal decrease in the zinc amalgam concentration. This indicates that the aqueous phase should control mass transfer throughout the run; no change of slope, corresponding to a change in the controlling phase, should be denoted in a plot of Eq. (6). Figure 3 shows the parameters of Eq. (6) plotted for the data obtained in run 146, A definite change of 26 ORNL DWG 75-4970Ri (.2 — 14— 1.0 08 - 08— 0.7 — ) (o) ; 0.6 |— /ot i) Cpp _gnK 0.5 0.4 — 0.3 — 0.2 — Ol i— 0.0 | 1 1 { | | ] ] 1 | | O 3 6 3 {2 15 18 24 24 27 30 TIME (min) Fig. 3. Logarithm of the concentration ratio vs time for ruan 146 in the water—-mercury contactor. 27 slope is recognizable at a value of the abscissa (time) between 15 and 13 min, which would indicate a change in the phase containing the limiting resistance to mass transfer. Mass transfer theory predicts that the value of the individual phase mass transfer coefficient is constant. The experimental results given in Table 4 indicate that the measured mass transfer coefficient for the mercury phase in runs 145, 148, and 149 increased in direct proportion to the initial lead ion concentration. Similarly, the values reported for the aqueous phase mass transfer coefficient in runs 146 and 147 vary inversely to the initial lead ion concentration. It is evident that for this system the model we have developed is inadequate, and further studies need to be made to completely understand the system. 6. FUEL RECONSTITUTION DEVELOPMENT: INSTALLATION OF EQUIPMENT FOR A TUEL RECONSTITUTION ENGINEERING EXPERIMENT R. M. Counce The reference flowsheet for processing the fuel salt from a molten- salt breeder reactor (MSBR) is based upon removal of uranium by fluorina- : . . 12 . . . tion to UFG as the first processing step. The uranium removed in this step must subsequently be returned to the fuel salt stream before it returns to the reactor. The method for recombining the uranium with the fuel carrier salt (reconstituting the fuel salt) is to absorb gaseous UF6 into a recycled fuel salt stream containing dissolved UF, by utilizing 4 the reaction: UFg oy + UF,qqy = 20F5qy - (16) The resultant UF5 would be reduced to UF4 with hydrogen in a separate vessel according to the reaction: + 1/2 H + HF ) (17) (2) Vs cay 2¢g) = Faqa) 28 We are beginning engineering studies of the fuel reconstitution step in order to provide the technology necessary for the design of larger equipment for recombining UF6 generated in fluorinators in the processing plant with the processed fuel salt returning to the reactor. During this report period, equipment that was described previously13 was fabricated and is being installed. This report contains documentation of equipment and adds detail to the previous report. The major components of the fuel reconstitution engineering experi- ment (FREE) are shown schematically in Fig. 4, and in move detail in Figs. 5 and 6. The equipment for this experiment consists of a 36-liter feed tank, a UF6 absorption vessel, a H2 reduction column, an effluent stream sampler, a 36~liter receiver, NaF traps for collecting excess U'F6 and disposing of HF, gas supplies for argon, hydrogen, and UF and means 63 for analyzing the gas streams from the reaction vessels. 6.1 Equipment Documentation The feed and receiver tanks are almost identical in construction, with the exception that the feed tank has an additional nozzle. The similarity of these vessels is seen in Figs. 7 and 8. The feed tank has five nozzles, as seen in Fig. 9, as follows: (1) a salt inlet nozzle (1/2-in. sched 40 pipe with a fitting for a sleeved 3/8-in. tube that does not extend into the tank) ; (2) a-salt exit nozzle (1/2-in. sched 40 pipe with a fitting for a sleeved 3/8-in. tube that extends to 1/2 in. of the bottom of the tank); (3) a sparge nozzle (3/8-in. sched 40 pipe with a fitting for a sleeved 1/4-in. tube extending to within 1/2 in. of the bottom of the tank); (4) an off-gas and pressurization nozzle (3/8-in. sched 40 pipe with a fitting for a sleeved 1/4-in. tube that does not extend into the tank); and (5) a sampler port (3/4-~in. pipe with a 3/4~-in. ball valve and cap). The return tank has nozzles 1 through 4 of the above, as seen in Fig. 10; however, the only pressurization capacity is through the sparge line. 29 ORNL DWG 74-1i666R! - - E ] E—= UFS HF UFG ANALYSIS ! |ANALYSIS HF TRAP ' TRAP o . e ‘ BUILDING OFF~(GAS Ufg . ;————’ SYSTEM b g } EFFLUENT STREAM * P * SAMPLER FREEZE ; i VALVES B Ar-——-—] ! t : UFG 1 : ABSORPTION 1 i VESSEL : : : | ! [ ' FEED TANK RECEIVER Hy L H, REDUCTION COLUMN Fig. 4. Schematic flow diagram for the fuel reconstitution engi- neering experiment (FREE). ORNL -DWG 7%-37R2 30 -+ N, - UFg - Hp —=~ VACUUM OFF-GAS - Ar Ha OFF-GAS B — HF TRAP RECEIVER TANK ABSORPTION VESSEL Ha ; REDUCTION | COLUMM TAMNK FEED EE. for FRE i gram 5. Flow disa Fig. 31 ORNL-DWG 75-35R2 Fi 101 1S A HASTINGS MASS FLOWMETER MODIFIED TO GIVE A 10-50 mA OUTPUT SIGNAL -— ARGON SYSTEM CALIBRATION TRAP UFg TRAP UFg TANK Fig. 6. UF6 generation and metering system' flow diagram for FREE. SR YRS S GAK RIDGE NATIONAL LABORATORY 3 4 5 6 7 8 2 Eie . ia Feed tank for FREE. PHOTO 3144-74 k PHOTO 3141-74 OAK RIDGE NATIONAL L ABORATORY i A # & 7 4 ¥ . " ‘ Receiver tank for FREE. ee PHOTO 3151-74 ™ i < [ \ % (5] wn PHOTO 3148-74 e OAK RIDGE ! ATIONAL LABORATORY ) f 2 3 4 4 6 & 8 | 9 \O | Fig. 10. Top view of receiver tank for FREE. 36 A side view of the UF6 absorption vessel is shown in Fig. 11, and a top view is shown in Fig. 12. The nozzles in Fig. 12 are identical with nozzles 1 through 4 of the return tank with one exception. The off-gas nozzle is a 3/8-in. sched 40 pipe with a 3/8-in. Swagelock fitting which serves as the off-gas line from that vessel. The H2 reduction column is shown in Figs. 13-15. Two nozzles are shown in a top view of the column (Fig. 14): (1) a salt inlet nozzle (as previously described for the feed tank); and (2) an off-gas nozzle (as described for the UF vessel). 6 absorption Figure 15 shows the 3/8-in.-pipe side arm (the gas inlet and distributor). The hydrogen supply line, a 1/4-in. tube, is welded to this side arm. Also shown is the 3/8-in.-pipe salt exit port. The salt exit line, a 3/8-in. tube, is welded to the exit port. The flowing stream sampler is shown in Fig. 16. Access to the liquid phase is obtained through the 3/4-in. ball valve. The sampler is equipped with a 3/8-in. tube fitting for an off-gas line. A 3/8-in. tube fitting for a "cover gas'" line is not shown. The UF, and HF traps are identical in construction except for the 6 length. The UF. trap is 55-5/8 in. long, while the HF trap is 31-5/8