M. J. Barnes, B. B. Anderson, T. L. White, and K. B. Martin
Westinghouse Savannah River Company
Aiken, SC 29808

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any 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. Reference herein to any specific commercial product, process or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

This report has been reproduced directly from the best available copy.

Available for sale to the public, in paper, from:  U.S. Department of Commerce, National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22161,  phone: (800) 553-6847,  fax: (703) 605-6900,  email:  orders@ntis.fedworld.gov   online ordering:  http://www.ntis.gov/support/ordering.htm

Available electronically at  http://www.osti.gov/bridge/

Available for a processing fee to U.S. Department of Energy and its contractors, in paper, from: U.S. Department of Energy, Office of Scientific and Technical Information, P.O. Box 62, Oak Ridge, TN 37831-0062,  phone: (865 ) 576-8401,  fax: (865) 576-5728,  email:  reports@adonis.osti.gov

This report describes batch testing of six different tank wastes for catalytic NaTPB decomposition. The waste samples came from Tanks 7F, 13H, 26F, 30H, 35H, and 46F. Tests with the six wastes occurred at both ambient (22-26 ° C) and elevated (45 ° C) temperature. Testing lasted six months. Samples obtained from the tests yielded the following observations.

• Tests conducted with Tank 35H waste produced detectable concentrations of triphenylborane, a NaTPB decomposition product.
• Under the conditions tested, the maximum reaction rate observed – for samples from Tank 35H -- equates to a theoretical benzene generation rate of 0.2 mg/(L-h).
• The maximum rate observed at 25 ° C fell well within the design basis limit for the proposed facility.
• Palladium concentrations in the six waste samples, when diluted to 4.7 M sodium, ranged from 0.05 to 0.16 mg/L. Analyses did not detect the presence of platinum in the waste.
• The waste sample from Tank 35H contained the highest concentration of soluble mercury, an element known to participate in the reaction sequence. This finding suggests the value of continuing the efforts to explore the influence of soluble mercury concentration on the catalytic reaction.

Keywords: Salt Waste Processing Facility (SWPF), Tetraphenylborate, Cesium, Catalyst

The Department of Energy discontinued operation of the In-Tank Precipitation facility due to the potential for catalytic decomposition of sodium tetraphenylborate. The Salt Disposition Systems Engineering Team identified Small Tank Tetraphenylborate Precipitation (STTP) as an alternative to replace the In-Tank Precipitation Facility at the Savannah River Site. The STTP process applies the same process chemistry for removal of cesium from the radioactive wastes but at a controlled lower temperature and in a smaller facility that offers engineering features to mitigate potential for a catalytic reaction. However, additional understanding of the catalytic reaction, through further experimental investigation, is needed to better define the potential for a reaction to occur in the proposed facility. Simulant testing indicates that the minimum species required to catalyze sodium tetraphenylborate (NaTPB) decomposition are either copper or a supported, reduced noble metal (e.g., Pd(0) on alumina). The most reactive simulant system includes diphenylmercury (or Hg(II) salt) along with the alumina supported Pd(0) and diphenylborinic acid (2PB). The most reactive radioactive system observed was from batch testing with actual high level waste in 1997-98.High Level Waste Engineering (HLWE) requested that the Savannah River Technology Center (SRTC) further evaluate the potential catalytic properties of additional high level waste tank samples.^,, This report provides the details of those experiments and partially addresses Item 2.0 (Cesium Removal Kinetics and Equilibrium) of the HLW Applied Technology Scope of Work Matrix for Small Tank Tetraphenylborate (Demonstration Phase).

Experiments tested the catalytic activity of six high level waste tank samples. The experiments occurred in remote cells using manipulators to perform test operations. The tank wastes used in the program provided varying process histories and composition. The samples were as follows:

• Tank 7F: F-Area Waste Removal Tank
• Tank 13H: H-Area Waste Removal Tank
• Tank 26F: F-Area Evaporator Feed Tank
• Tank 30H: H-Area Evaporator Receipt Tank
• Tank 35H: H-Area Canyon Receipt Tank
• Tank 46F: F-Area Concentrate Receipt Tank

Analysis of filtered and unfiltered aliquots of the samples by inductively coupled plasma mass spectroscopy (ICP-MS), inductively coupled plasma emission spectroscopy (ICP-ES), plutonium scintillation with thenoyl trifluoroacetone (PuTTA), ion chromatography (IC), titration, atomic absorption (AA), scintillation, and gamma scan yielded both soluble and insoluble compositions of the waste. Appendix 1 provides the results of the characterizations.

Each waste solution was prepared for testing by diluting its entire sample volume to 5.05 M Na⁺. The actual volume of each solution, after dilution, was unique due to their varying starting volumes and sodium concentrations. The experiments used custom designed 250-mL stainless steel vessels that permitted oxygen depletion (see Figure 1). Each vessel was fitted with two gas purge ports (hose adapters and ball valves) as well as a septum sample port and sparge tube assembly port. Each sample was split in half. The twelve resulting fractions were placed in individual specially built vessels.

Figure 1. Custom designed 250-mL stainless steel reaction
vessels with nitrogen sparge tubes.

As a prerequisite to catalytic testing, the authors evaluated the conditions required to produce a deaerated solution. We performed a sparge test in one of the unique vessels with a stainless steel frit sparge tube. An Ocean Optics FOXY fiber optic oxygen sensor was used to monitor oxygen depletion during nitrogen sparge at a fixed flow rate. The sensor was initially calibrated in a large (1 L) stainless steel baffled vessel with impeller and air or nitrogen sparge at room temperature. The sparge rates were kept reasonable so as not to cool the liquid substantially. The sensor was then placed in the small stainless vessel, with no agitation, containing 150 mL deionized water. Gas flow was controlled via a rotameter and monitored with a Humonics flow meter (0-500 standards cubic centimeters per minutes (sccm) flow range). Six experiments were performed over several days in order to assess the variability of the measurement scheme. The target flow rate was 200 sccm, and this varied during the course of the development testing with a standard deviation of 34 sccm (several runs contained a large drift component due to the instability in the rotameter). Initial conditions were reached by flowing air through the vessel until the sensor reached steady state at approximately 22 vol % oxygen. At this point, the system was switched to nitrogen and adjusted to 200 sccm flow. The data were acquired at 30 s and 5 s increments, but the FOXY software recorded the hour and minute values, leaving the second values as zeros. Therefore, the time axis had to be interpolated, and this was accomplished by applying a smoothing filter to the time axis, which essentially generated increments smaller than a minute. The data were then analyzed via Microsoft Excel calculations to find the least squares fit of the depletion curve to an exponential function. From this fit, time constants were obtained for each run. The data are shown in Table 1 and Figure 2.

Table 1. Sparge test results for 150 mL liquid volume, nominal 200 sccm flow rate,
sparge time constants determined using FOXY sensor to monitor oxygen depletion.

Experiment number	Flow rate at end of run, sccm	Exponential time constant, seconds
1	223.0	259.9
2	200.0	187.1
3	288.0	205.7
4	201.0	260.9
5	200.0	258.0
6	208.0	493.4
Average	220.0	277.5
std. dev.	34.5	110.4

The data show a time constant for the depletion of oxygen from the system of 277 seconds. Three times the time constant yielded an approximate sparge time for the conditions of 832 seconds, or 14 minutes. The standard deviation on this number was 330 seconds (3 times the standard deviation of the runs), or approximately 6 minutes. Therefore, a minimum recommended sparge time for the experiment was 20 minutes to accomplish oxygen depletion by nitrogen sparging at 200 sccm.

Based upon the simulant sparge test findings, each test solution was conservatively sparged for 180 minutes. The sparge tubes were removed immediately upon completion of the deaeration step and the port capped. The vapor space of the vessels was then purged with nitrogen to complete the vessel inerting process.

Each vessel was charged with 0.55 M NaTPB solution via syringe through the septum port. The quantity of NaTPB added was sufficient to react stoichiometrically with the measured potassium and provide a 0.03 M residual excess of NaTPB. The percent excess added for each tank is shown in Table 2. Additionally, each vessel was charged with 500 mg/L of phenylboronic acid (1PB) to promote reactivity and alleviate the lengthy induction periods observed in the previous batch tests with high level waste. The resulting target sodium concentration for each test after the additions was 4.7 molar. Six test vessels, representing the six waste samples, were placed in a heater-shaker

Figure 2. Sparge test results for 150 mL liquid volume, nominal 200 sccm flow rate, sparge time
constants determined using FOXY sensor to monitor oxygen depletion.

Table 2. Percent Excess NaTPB added to each real waste batch test.

where they were shaken continuously at ~190 rpm and heated at 45 ± 3 ° C. The six test vessels were designated with the letter H to indicate they were the higher temperature of the two sets (e.g., Test 7F-H, 13H-H, etc.). The remaining six test vessels were labeled with the letter L (e.g., Test 7F-L, 13H-L, etc.) to indicate they were the lower temperature set. These six were also shaken continuously but no temperature control was provided. Ambient temperature in the shielded cell facility ranged from 28 ° C (start of testing) to 22 ° C (end of testing). Figure 3 is a photograph of the two shakers and test vessels in the shielded cell. Testing lasted a period of 6 months. Periodically, samples were obtained via syringe, filtered immediately using 0.45 micron nylon disposable filter cups, and the filtrate typically submitted for Cs-137 (gamma scan), phenylborates (High Performance Liquid Chromatography, HPLC), and soluble boron (microwave digestion followed by ICP-ES) analysis. The available test volume and previous sample data dictated the sample frequency and required analysis. Data for the twelve tests are contained in Appendix 2.

Figure 3. Photograph of both the heater-shaker and ambient temperature
shaker in operation in the shielded cell.

Cesium decontamination of the test solutions proved erratic between the twelve tests. Five of the tests (i.e., 7F-L, 26F-H, 35H-L, 35H-H, and 46F-H) exhibited rapid cesium precipitation, reaching the Saltstone limit of 45 nCi/g in a few hundred hours. Three tests (7F-H, 26F-L, and 30H-H) precipitated cesium at a slower rate, reaching the Saltstone limit around a thousand hours of testing. The remaining four tests (13H-L, 13H-H, 30H-L, and 46F-L) took considerably longer. Figure 4 shows the three distinct types of cesium precipitation behavior. The lack of any pattern between tank number, temperature, or even percent excess NaTPB of the affected tests indicates that another property affected the rate of decontamination. The most probable cause is the extremely low solubility of NaTPB in high ionic strength salt solutions such as these tested. Once added, the majority of the NaTPB precipitated out and its dissolution was very slow under the mild degree of agitation. An attempt was made to increase the shaker speeds but this resulted in breaking of the clamps holding the vessels in place. The decision was made to add additional NaTPB to the tests. However, the available material from Test 13H-H was depleted earlier than expected. As a result, no additional NaTPB was added to the test and it never obtained decontamination. Additional NaTPB was not added to Test 46F-L since it exhibited signs of increasing decontamination. Test 30H-L achieved decontamination soon after the second addition of NaTPB. Test 13H-L showed only a moderate level of decontamination and failed to reach the Saltstone limit. Interestingly, the cesium data from Test 13H-L does not rule out the possibility that reaction of solid tetraphenylborate was occurring at the end of the test (however, the soluble boron data does not verify this observation). It should be noted that the presence of soluble cesium indicates the amount of soluble NaTPB is very low and hence reaction of soluble NaTPB in the system is limited. However, the long test duration magnifies detection of the even the slowest of reactions and offers ample time for overcoming induction periods or long catalyst activation times. Of the twelve tests, seven showed measurable levels of soluble NaTPB. These were Tests 7F-L, 7F-H, 26F-L, 26F-H, 35H-L, 35H-H, and 46F-H.

Figure 4. Graph exhibiting the three distinct types of
cesium precipitation behavior exhibited by the twelve tests.

Evaluation of High-Level Waste Test Data for Catalyzed Decomposition of NaTPB

Eleven of the twelve tests (Test 13H-H material depleted too quickly to perform HPLC analysis) showed decomposition of the added 1PB and the resulting formation of nearly equivalent amounts of phenol. This demonstrates that the solutions are capable of producing phenylborate decomposition. However, the formation of triphenylborane (3PB) and to some extent 2PB provides the most direct evidence of catalytic NaTPB decomposition. Only tests 35H-L and 35H-H showed detectable quantities of 3PB. It is feasible that a reaction might occur which produced no detectable 3PB, 2PB, or 1PB (i.e., 3PB would form at a slow rate and then be consumed along with any 2PB and 1PB at a sufficiently fast rate so that no 3PB, 2PB, or 1PB would be detected). Test 35H-H also produced detectable quantities of 2PB.

Figure 5 provides the reaction profile for Test 35H-H. For comparative purposes, the reaction profile of Test 26F-H, a non-reactive case, is shown in Figure 6. The rate of increase in soluble boron from Test 35H-H conservatively indicates the benzene generation rate from NaTPB decomposition to be less than 0.2 mg/(L· h). A similar analysis of Test 35H-L yields a conservative rate of benzene generation of less than 0.08 mg/(L· h). Note that the calculated rates are based upon small changes in soluble boron concentration and are conservative . The actual values are likely less than indicated. In consideration, such small changes in soluble boron are evident in other tests (e.g., Test 30H-H). However, Tests 35H-L and Test 35H-H differ from the others in that the boron data shows a more consistent trend, rather than scatter, and reaction did occur as evidenced by formation of 3PB and 2PB (not observed elsewhere). Regression of the data from Test 35H-H yields rate constants of 3.5 E-5, 3.3 E-4, 2.9 E-3, and 5.8 E-3 h^-1 for NaTPB, 3PB, 2PB, and 1PB, respectively. These rates are of similar magnitude to those obtained from the decomposition of the same species in Tank 49H in January 2000. Rate constants were not obtainable from the Test 35H-L data. The low benzene generation rate observed in Test 35H-H is more than 2 orders of magnitude less than observed (~ 40 mg/(L· h) in the previous demonstration with high level waste. Note, that the latter rate serves to define the baseline reaction rate for the facility design.

A comparison of the composition of Tank 35 waste (post-dilution) versus the other five tank wastes yields some insight into its greater (relative) degree of reactivity. Most notably, the soluble mercury level in the Tank 35H waste is greater than in any other tank waste sample (seeTable 3). Interestingly, Tank 46F waste sample contains more total mercury, but it is largely insoluble (most likely insoluble HgO). Examination of the Pd concentration (also post-dilution –

Figure 5. Reaction profile of Test 35H-H, a reactive test case.

Figure 6. Reaction profile of Test 26F-H, a non-reactive test case.

see Table 4) does not reveal any significant observations. All tests had measurable Pd concentration. In almost all cases the majority was soluble in the as received samples. The concentration of Pd in the tests was remarkably close and ranged from 0.05 to 0.16 mg/L. Therefore, all should have approximately the same level of reactivity if all of the Pd in each case were able to activate. A Pd concentration of 0.2 mg/L approximates the most observed in previous real waste tests with significant reactivity. The lack of reactivity suggests that another species (e.g., mercury) also participates in the reaction process. No other noble metal species stands out in Tank 35H waste (in comparison to the other waste samples – see Appendix 1).

Table 3. Mercury concentrations of the six waste
tank samples used in testing (after dilution to 4.7 M).

Table 4. Palladium concentrations of the six waste
tank samples used in testing (after dilution to 4.7 M).

This report describes batch testing of six different tank wastes for catalytic NaTPB decomposition. The waste samples came from Tanks 7F, 13H, 26F, 30H, 35H, and 46F. Tests with the six wastes occurred at both ambient (22-26 ° C) and elevated (45 ° C) temperature. Testing lasted six months. Samples obtained from the tests yielded the following observations.

• Tests conducted with Tank 35H waste produced detectable concentrations of triphenylborane, a NaTPB decomposition product.
• Under the conditions tested, the maximum reaction rate observed – for samples from Tank 35H -- equates to a theoretical benzene generation rate of 0.2 mg/(L-h).
• The maximum rate observed at 25 ° C fell well within the design basis limit for the proposed facility.
• Palladium concentrations in the six waste samples, when diluted to 4.7 M sodium, ranged from 0.05 to 0.16 mg/L. Analyses did not detect the presence of platinum in the waste.
• The waste sample from Tank 35H contained the highest concentration of soluble mercury, an element known to participate in the reaction sequence. This finding suggests the value of continuing the efforts to explore the influence of soluble mercury concentration on the catalytic reaction.

Additional real waste batch tests are warranted if Small Tank Tetraphenylborate Precipitation is selected as the technology for deployment in the Salt Processing Facility. Specifically, additional testing with Tank 35H waste using increased agitation would likely provide more reliable reaction rates. Furthermore, continued batch testing with additional tank waste samples and increased agitation rates would serve to increase the available data sets investigating the reactivity of high level waste.

Personnel utilized reagent grade chemicals and prepared solutions with calibrated balances checked daily before use. The weights used for balance checks received calibration by the SRTC Standards Laboratory. The accuracy of glassware used to measure volumes was verified by gravimetric methods using water as a standard. All M&TE used in this task received calibration or verification for accuracy prior to their use. The Analytical Development Section performed all chemical analyses. Laboratory Notebook WSRC-NB-2000-00125 (M. Barnes) contains the experimental log for these experiments. Anderson recorded the data from examination of the nitrogen sparging efficiency in Laboratory Notebook WSRC-NB-2000-00214. This report, in part, finalizes the work identified by Item 2.0 of "Applied Technology Integration Scope of Work Matrix for Small Tank TPB Precipitation (Demonstration Phase)," HLW-SDT-99-0353, Rev. 5, November 21, 2000.

The authors thank the members of the various support groups that contributed to the success of this project. The work benefited from the efforts of the talented staff that provided analytical support for the program. Specifically, Annie Still performed the high-pressure liquid chromatography, Cici Diprete supervised the gamma spectroscopy, Beverly Burch performed the chemical digestions of samples to prepare for boron analysis, June Hart supervised the ICP-ES analysis for soluble boron, and Bill Boyce supervised the ICP-MS work. Debbie Sanders, Dee Wheeler, Phyllis Burkhalter, and Kim Prettel each served as Technical Analysts for various tasks within the overall effort. The authors wish to extend a debt of gratitude to Sam Fink for the tireless hours he spends making sure all test activities and personnel activities run smoothly. Lastly, an effort of this magnitude succeeds due to the efforts of so many individuals. We apologize to those we forgot to mention in this list.

References

1. D. D. Walker, M. J. Barnes, C. L. Crawford, R. F. Swingle, R. A. Peterson, M. S. Hay, and S. D. Fink, "Decomposition of Tetraphenylborate in Tank 48H (U)", WSRC-TR-96-0113, Rev. 0, May 10, 1996.
   
2. P. V. Bonnessen, L. H. Delmau, and F. V. Sloop, Jr., "Investigating the Mechanism of Catalytic Tetraphenylborate Decomposition Using Nuclear magnetic Resonance Spectrometry: Initial studies in FY00", ORNL/TM-2000/279, December 2000.
   
3. M. J. Barnes and L. N. Oji, "Small Tank Tetraphenylborate Catalyst Studies", WSRC-TR-2001-000180, March 28, 2001.
   
4. D. T. Hobbs, M. J. Barnes, R. A. Peterson, and C. L. Crawford, "Radioactive Testing Results in Support of the In-Tank Precipitation Facility", WSRC-TR-98-00070, April 2, 1998.
   
5. P. L. Rutland, HLW-SDT-TTR-99-0020, Rev. 0, December 1999.
   
6. P. L. Rutland, HLW-SDT-TTR-99-0021, Rev. 0, December 1999.
   
7. P. L. Rutland, HLW-SDT-TTR-99-0027, Rev. 0, December 1999.
   
8. Salt Disposition Systems Engineering Team, "Applied Technology Integration Scope of Work Matrix for Small Tank Tetraphenylborate Precipitation (Demonstration Phase)", HLW-SDT-99-0353, Rev. 5, December, 2000.
   
9. Bases, Assumptions, and Results of the Flowsheet Calculations for the Decision Phase Salt Disposition Alternatives, WSRC-RP-99-00006, Rev. 2, p. 59, April 2001.
   
10. D. D. Walker, "Balance Check" (U), Manual L12.1, Procedure IWT-OP-011, Rev. 3, December 6, 1994.
    
11. D. D. Walker, "Calibration of Laboratory Glassware" (U), Manual L12.1, Procedure IWT-OP-009, Rev. 3, May 8, 1995.