Tetraphenylborate Catalyst Development for the Oak Ridge National
Laboratory 20-L Continuously Stirred Tank
Reactor Demonstration

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

Keywords: Sodium Tetraphenylborate, Catalyst, Salt Processing Facility

Prior to testing in the 20-L facility at Oak Ridge National Laboratory, the program required smaller (1 L) bench-scale tests to ensure the reactivity of the catalyst system approximated the target benzene generation rate of 10 mg/(L· h) under continuous operation. Information contained in this document details these tests and the refinement of the catalyst system. High Level Waste (HLW) Process Engineering requested that the Savannah River Technology Center (SRTC) perform this work via Task Technical Requests HLW-SDT-TTR-99-0020, HLW-SDT-TTR-99-0021, and HLW-SDT-TTR-99-0027.^9,10,11 This work 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).¹²

Figure 1. Bench-scale CSTR and Concentrate Tank Set-up.

Figure 2. Close-up of Bench-scale CSTR and Concentrate Tank.

Figure 3. Concentrate Tank lid assembly with downcomers and impeller after DF Test 1.

Results and Discussion

Test 1: DF Test at 45°C (No Catalyst)

The initial demonstration of the set-up, shown in Figure 1, occurred in the absence of catalyst. This tests demonstrated precipitation performance under a given set of conditions, specifically at 45°C and with IIT B-52 antifoam added. Table 2 and Table 3 contain the potassium data from the test. Figure 4 shows the soluble potassium concentrations for each vessel. The CSTR obtained DF in excess of 1000 in approximately 13 hours while the Concentrate Tank required 28 hours to reach maximum reported DF. In the test, DF’s ranging from 1100 to greater than 3000 occurred in the CSTR while those in the Concentrate Tank exceeded 3000. The cause of the observed fluctuation in the CSTR data remains unknown. The test demonstrates adequacy of the equipment design and test method and further indicates that the antifoam does not detract from the precipitation kinetics.

Figure 4. Potassium data from initial, no-catalyst DF Test 1 at 45°C.

Test 2: 26 mg/L Pd at 45°C

Test 2 repeated the conditions of Test 1 except with catalyst present. Personnel heated the vessels and maintained them at 45°C (± 1°C). Researchers introduced a post-dilution concentration of 26 mg/L Pd(0) (on alumina) to the CSTR. The test operated for 72 hours during which time the tetraphenylborate solids of the Concentrate tank concentrated from 0.5 wt % to approximately 2.6 wt (as calculated by material balance). As a result, the insoluble Pd catalyst also concentrated a factor of five times. Table 4 and Table 5 contain data from the test. Figure 5 displays the decontamination behavior. Decontamination factors ranging from 1300 to 2700 occurred in the CSTR. The Concentrate Tank obtained a DF of 1700, which then declined. Examination of Figure 6 and Figure 7 reveals a significant tetraphenylborate reaction occurring in both vessels as evidenced by the production of both triphenylborane (3PB) and diphenylborinic acid (2PB). As expected, the 1PB concentration shows a steady growth from the start of the reaction and then levels near 500 mg/L. Note that reaction did not occur until after the vessel achieved decontamination yielding soluble NaTPB. The reaction proved sufficient to consume all soluble NaTPB. The authors used changes in the phenylborate species (excluding 1PB) to calculate an approximate benzene generation rate. For simplicity, we did not use the 1PB data since we added this material to the test. Figure 8 contains a plot of approximate benzene generation rate. The calculated values underestimate the actual benzene generation rate since they do not account for formation of 1PB from 2PB and borate from 1PB. The calculated rate can also mislead the reader since the Concentrate Tank quickly became exhausted of NaTPB, 3PB, and 2PB. In summary, benzene generation rates in excess of 30 mg/L occurred. We believe the actual rate proved significantly higher. The high rate indicated the need of further studies at lower temperatures to achieve the desired conditions – 10 mg/(L· h) in a 10 wt % solids slurry – for the ORNL testing.

Figure 5. Potassium data from Test 2 with 26 mg/L Pd(0) at 45°C.

Figure 6. CSTR phenylborate data from Test 2 with 26 mg/L Pd(0) at 45°C

Figure 7. Concentrate Tank phenylborate data from Test 2 with 26 mg/L Pd(0) at 45°C.

Figure 8. Approximate benzene generation rate from Test 2 with 26 mg/L Pd(0) at 45°C.

Test 3: 26 mg/L Pd at 35°C

Test 3 used identical conditions as in Test 2 except maintaining the temperature at 35°C. The test ran for 72 hours without incident. Figure 9 shows that the precipitation behavior of the CSTR and Concentrate Tank behaved well. Both vessels exhibited decontamination factors in excess of 3000. Furthermore, the Concentrate Tank did not lose DF as observed in Test 2. Table 6 and Table 7 contain the data for the test. Examination of the phenylborate data shown in Figure 10 and Figure 11 indicates a fast reaction occurred. Unlike Test 2, soluble NaTPB existed in the Concentrate Tank throughout the test duration. Figure 12 shows the approximate benzene generation rate as a function of time. The calculated benzene generation rate in the CSTR exceeded 15 mg/(L· h) at the end of testing. In the Concentrate Tank, the benzene generation rate approximated 5 mg/(L· h) at 2.6 wt % tetraphenylborate solids. The decreased rate may be due in part to either catalyst deactivation, excessive reactivity in the CSTR (reduced available NaTPB), or underestimation since the method used to calculate the rate ignores 1PB and borate concentration changes. Again, the fast CSTR reaction warranted further testing at less reactive conditions, specifically 25°C.

Figure 9. Potassium data from Test 3 with 26 mg/L Pd(0) at 35°C.

Figure 10. CSTR phenylborates data from Test 3 with 26 mg/L Pd(0) at 35°C.

Figure 11. Concentrate Tank phenylborates data from Test 3 with 26 mg/L Pd(0) at 35°C.

Figure 12. Approximate benzene generation rate from Test 3 with 26 mg/L Pd(0) at 35°C.

Test 4: 26 mg/L Pd at 25°C

Test 4 used identical conditions as in Tests 2 and 3 except maintaining the temperature at 25°C. Figure 13 shows the precipitation behavior of Test 4 while Table 8 and Table 9 contain the potassium and phenylborate data. The decontamination behavior of Test 4 proved similar to that of Test 3. Decontamination factors in excess of 3000 resulted for both vessels. Additionally, soluble NaTPB existed in both vessels at the end of testing. Examination of Figure 14 and Figure 15 show a large soluble concentration of NaTPB early in testing. This high concentration resulted from production of a more dilute salt waste simulant in the vessels than targeted. As evidenced by the data, personnel quickly observed and corrected for the flow imbalance. This action produced a decrease in the soluble NaTPB concentration. Again, reaction occurred as evidenced by the formation of 3PB and 2PB. Figure 16 provides a graph of the approximate benzene generation rate. Unlike Tests 2 and 3 which exhibited continual growth of the benzene generation rate even in the CSTR, Test 4 exhibits a maximum benzene generation rate in the CSTR of less than (i.e., detection limited) 5 mg/(L· h). The benzene generation rate in the Concentrate Tank appears to level around 12 mg/(L· h). Testing at ORNL will need to use a slightly reduced catalyst concentration to produce the desired benzene generation rate in a more concentrated system. The Test 5 designed attempted to achieve the required concentration.

Figure 13. Potassium data from Test 4 with 26 mg/L Pd(0) at 25°C.

Figure 14. CSTR phenylborates data from Test 4 with 26 mg/L Pd(0) at 25°C.

Figure 15. Concentrate Tank phenylborates data from Test 4 with 26 mg/L Pd(0) at 25°C.

Figure 16. Approximate benzene generation rate from Test 4 with 26 mg/L Pd(0) at 25°C.

Test 5: 7.8 mg/L Pd at 25°C

Test 5 repeated the conditions of Test 4 but with the Pd concentration dropped from 26 mg/L to 7.8 mg/L. Additionally, Test 5 continued for a longer duration (200 h) to allow the contents of the Concentrate Tank to reach 7 wt %. Figure 17 shows the precipitation behavior of the two vessels. Potassium DFs of approximately 1500 or greater occurred. Near the end of the test, both vessels experienced a decline in DF due to a decrease in NaTPB feed. Table 10 and Table 11 contain the data. Figure 18 and Figure 19 show the phenylborates concentration data. Similar to Test 4, the soluble NaTPB proved higher than expected early in the test. Again this increased concentration occurred due to over-dilution of the salt simulant. Personnel quickly corrected flows and NaTPB concentrations returned to near the expected level. Researchers removed 1PB from the feed solution at approximately 150 h to determine if the reaction persisted without this additive. A maximum approximate benzene generation rate of less than 5 mg/(L· h) resulted in the CSTR at 0.5 wt % tetraphenylborate solids (see Figure 20). The benzene generation rate in the Concentrate Tank showed continual increase to approximately 10 mg/(L· h) over the first 150 hours prior to removal of the 1PB (also see Figure 20). The rate appeared to level after that point. For comparative purposes, the benzene generation rate was plotted versus the total Pd accumulated in the Concentrate Tank for both Tests 4 and 5 at 25°C (see Figure 21). The total Pd accumulated in the Concentrate Tank during each test was approximately the same due to offsetting variables. The Pd feed concentration was approximately three times higher in Test 4 than in Test 5. However, the test duration for Test 5 was almost 3 times longer than in Test 4. Data for the two tests are in agreement for the bulk of the testing. Only in Test 5 at the end of the test does the benzene generation show slight deviation from Test 4. Examination of the NaTPB feed rate data for Test 5 indicates that over the last 48 hours of testing, significantly less NaTPB entered than desired (shown graphically in Figure 22 through Figure 24). This was attributed to deviation in flow rate caused by variation of the pump rate. This explains the loss of DF observed in Figure 17 at the end of testing. Test data suggests that the system requires a molar ratio of NaTPB to potassium of about 1.3, below which, the process did not maintain DF.

When personnel disassembled the equipment for cleaning, they observed a white solid residue adhered to the vessel walls and internal parts (see Figure 25). Characterization of the residue by SEM and EDS indicates the solids are composed of a mixture of aluminum and silicon particles embedded in a potassium rich matrix. Figure 26 shows the back-scattered image of the sample as well as the elemental scanning of the selected spots on the sample. From Figure 26, the majority of the material (based on volume) is made of a potassium rich matrix. This material is composed of elements lighter than sodium (the heaviest element detected by EDS). The researcher submitted the sample for molecular spectroscopy scanning. Looking at Figure 27, the spectrum of the "residue" shows absorption peaks shape and position characteristics of phenylborates (746, 716, 3000 and 3060 cm^-1). The residue spectrum did not contain any spectral feature of the antifoam. The residue did not contain degradation products from the antifoam. This is clearly shown in EDS spectrum of the residue (Figure 26) where there is not elemental sulfur (which is present in the antifoam). Looking at the spectrum of the antifoam (also contained in Figure 27), it is composed of an (long) aliphatic ester (1742, 2875, 2929, 2959 and 1460 cm^-1) and S(OH)₃ group as denoted by the 1256 and 1245 cm^-1 bands. The sulfate group is not ionized since there is a broad band at 3468 cm^-1 due to hydroxyl stretching. The residue is a mixture of inorganic and organic material. A closer examination by infrared spectroscopy of the silicon and aluminum elements detected by EDS revealed the presence of a mixture of silicon and aluminum hydroxide (recall the Pd catalyst was supported on alumina). Figure 28 shows the infrared spectrum of the CSTR residue by ATR (Attenuated Total Refluctance - Thunderdome). The technique is based on pressing the sample (about 1 mg) against a crystal. Pressing the sample against the crystal allows the infrared beam within the crystal to penetrate the sample. Since minerals (or most inorganics) have a high modulus, they offer a significant resistance to compression in the ATR experiment. Therefore, the inorganic portion of a mixture will preferentially wet the crystal and its corresponding infrared signature is detected. The inorganic material is amorphous as indicated by the X-ray diffraction (XRD) spectrum shown in Figure 29 where potassium tetraphenylborate is the only detected species. The majority of the residue is organic in nature.

Figure 17. Potassium data from Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 18. CSTR phenylborates data from Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 19. Concentrate Tank phenylborates data from Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 20. Approximate benzene generation rate from Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 21. Approximate benzene generation rate in the Concentrate tank as a function of total Pd from Tests 4 and 5 at 25°C.

Figure 22. NaTPB feed rate performance for Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 23. Feed and filtrate volumes processed during Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 24. Ratio of NaTPB to potassium for Test 5 with 7.8 mg/L pd(0) at 25°C.

Figure 25. Residue observed on CSTR impeller after Test 5 with 7.8 mg/L Pd(0) at 25°C.

Figure 26. SEM image of the residue found in Concentrate Tank from Test 5.
The figure includes EDS spectrum of selected spots in the SEM picture.

Figure 27. The infrared spectrum of the IIT B-52 antifoam agent used in this work and the residue found in the CSTR.

Figure 28. The ATR spectrum of the CSTR residue.

Figure 29. The XRD spectrum of the CSTR residue.

Data Tables

Table 1. Target concentration of salt waste simulant.

Table 2. CSTR data from DF Test 1 with no catalyst.

Table 3. Concentrate Tank data from DF Test 1 with no catalyst.

Table 4. CSTR data from Test 2 with 26 mg/L Pd(0) at 45°C.

Table 5. Concentrate Tank data from Test 2 with 26 mg/L Pd(0) at 45°C.

*ND indicates no data available.

Table 6. CSTR data from Test 3 with 26 mg/L Pd(0) at 35°C.

Table 7. Concentrate Tank data from Test 3 with 26 mg/L Pd(0) at 35°C.

Table 8. CSTR data from Test 4 with 26 mg/L Pd(0) at 25°C.

Table 9. Concentrate Tank data from Test 4 with 26 mg/L Pd(0) at 25°C.

Table 10. CSTR data from Test 5 with 7.8 mg/L Pd(0) at 25°C.

Table 11. Concentrate Tank data from Test 5 with 7.8 mg/L Pd(0) at 25°C.

*ND indicates no data obtained.

References

1. Bases, Assumptions, and Results of the Flowsheet Calculations for the Decision Phase Alternatives, WSRC-RP-99-00006, Rev. 0, p. 40.
2. 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.
3. 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.
4. R. A. Peterson and J. O. Burgess, "The Demonstration of Continuous Stirred Tank Reactor Operations with High Level waste", WSRC-TR-99-00345, Rev. 1, February 15, 2000.
5. D. D. Lee and J. L. Collins, "Continuous Flow Stirred Tank Reactor 20 Liter Demonstration Test: Final Report", ORNL/TM-1999/234, February 2000.
6. M. J. Barnes, "Catalyst Requirements for ORNL 20-L CSTR Demonstration (Test 2)", SRT-LWP-99-00099, Rev. 1, July 23, 1999.
7. R. A. Peterson and M. J. Barnes, "Results of Experimental Investigation into Noble Metal Catalyzed Decomposition of Sodium Tetraphenylborate", WSRC-TR-2000-00230, July 6, 2000.
8. L. N. Oji and M. J. Barnes, "Batch Studies of Sodium Tetraphenylborate Decomposition on Reduced Palladium Catalyst," WSRC-TR-2000-00459, October 28,2000.
9. P. L. Rutland, HLW-SDT-TTR-99-0020, Rev. 0, December 1999.
10. P. L. Rutland, HLW-SDT-TTR-99-0021, Rev. 0, December 1999.
11. P. L. Rutland, HLW-SDT-TTR-99-0027, Rev. 0, December 1999.
12. Salt Disposition Systems Engineering Team, "Applied Technology Integration Scope of Work Matrix for Small Tank Tetraphenylborate Precipitation (Demonstration Phase)", HLW-SDT-99-0353, December, 1999.