N. M. Hassan, W. D. King, D. J. McCabe, and L. L. Hamm
Westinghouse Savannah River Company
Aiken SC 29808

M. E. Johnson
CH2MHILL Hanford Group, Inc.
Richland, WA 99352

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

SuperLig^Ò 639 ion exchange resin is currently being evaluated for technetium (⁹⁹Tc) removal from radioactive Hanford tank wastes. To assess the performance of the resin in column configuration, a multiple batch contact method was used to determine the equilibrium distribution coefficients (K_d values) and percent removal for technetium (as pertechnetate, TcO₄^-) from highly alkaline waste solutions obtained from the Hanford Site. The K_d values for Hanford tank wastes AN-103, AZ-102, and AN-102 were 530, 886, and 287 mL/g, respectively. The TcO₄^- K_d values for the AN-102 mixed with AN-103 after 4 hours and 14 days were 456 and 392 mL/g, respectively. The percent removal of technetium as TcO₄^- anion from three Hanford tank solutions (AN-103, AZ-102, and AN-102) was higher than 80%, which is required for Hanford waste pretreatment flowshee. The results indicate that SuperLig^Ò 639 resin is highly selective for pertechnetate in the presence of relatively high concentrations of nitrate anion.

Millions of gallons of high-level radioactive waste (a legacy by-product of the production of nuclear materials for national defense needs during the past 50 years) are stored in underground tanks in the U. S. Department of Energy’s (DOE) Hanford Site awaiting pretreatment and safe disposal. Numerous processes involving the addition of many chemical reagents resulted in the generation of various waste streams, which were combined in underground storage tanks. The tanks contain varying amounts and types of solids and millions of gallons of aqueous supernate. While all of the supernate is alkaline and contains primarily sodium salts, significant compositional variation is observed between tanks. The major radioactive constituents in the waste are ¹³⁷C_S (t_1/2 = 30 y), ⁹⁹Tc (t_1/2 = 2 x10⁶ y), and ⁹⁰Sr (t_1/2 = 29 y), which are primarily responsible for the radioactivity due to their long half-lives and high fission yields. Technetium is primarily present in the supernate as pertechnetate (TcO₄^-), which is highly soluble in water and readily mobile in the environment [1,2].

The process flowsheet proposed for pretreatment of Hanford tank wastes removes ¹³⁷Cs, ⁹⁹Tc (as pertechnetate ion, TcO₄^-), and ⁹⁰Sr from the bulk of the waste and concentrates these species into smaller volume streams. The bulk of the decontaminated waste is to be vitrified and disposed as low-activity waste glass. The concentrated radionuclides are to be combined with high-level waste sludge and vitrified into high-level waste glass. To achieve this goal, novel process schemes and suitable ion exchange technologies are continuing to be developed and tested.

Technetium removal from Hanford waste supernates by ion exchange has been widely investigated [3-6]. Various ion exchange materials have shown the ability to remove ⁹⁹Tc as TcO₄^- from simulated waste solutions [7-10]. A highly-selective ion exchange resin, developed using molecular recognition technology and macrocyclic chemistry, was very effective at removing pertechnetate from highly alkaline Hanford waste solutions [11]. This resin, designated as SuperLig^Ò 639, is currently under consideration as the baseline ion exchange resin for the Hanford Waste Treatment Plant because of its high selectivity for pertechnetate, elutability, and availability in an engineered form that is suitable for use in ion exchange columns.

In support of the flowsheet development for the Hanford Waste Treatment Plant, multiple batch contact tests were performed with SuperLig^Ò 639 resin and actual Hanford tank waste supernates. The multiple batch contact method was employed to obtain equilibrium distribution coefficients (K_d’s) and percent removal (% R) for technetium from tanks 241-AN-103, 241-AZ-102, and 241-AN-102. In addition, the K_d’s were obtained for solution mixtures of tank 241-AN-103 with 241-AZ-102 and 241-AN-102 as well as for tank 241-AZ-102 with 241-AN-102, after aging the mixed samples for 4 hours and 14 days. The resulting data provided a rapid and cost-effective evaluation of this ion exchange material for technetium removal from complex tank waste solutions and process mixtures.

Radioactive waste samples from Hanford underground tanks were used for the determination of technetium batch distribution coefficients and percent removal. The samples were retrieved from Hanford tanks 241-AN-103, 241-AZ-102, and 241-AN-102. The prefix "241" is common to all Hanford tanks and will not be used hereafter. The primary compositions of the Hanford tank waste supernates are shown in Table 1. Tank AN-103 waste is characterized by having high alkalinity and high ionic strength. The "as-received" AN-103 sample was diluted to ~5 M Na⁺ with 0.01 M sodium hydroxide solution and filtered through 0.45-m m nylon filters prior to batch contact testing. The nitrate, free hydroxide, and ⁹⁹Tc concentrations in the diluted sample were 1.56 M, 1.87 M, and 0.052 mCi/mL, respectively. The "as-received" AZ-102 sample was relatively low in ionic strength (2.65 M Na⁺) and was not diluted. The sample was filtered through 0.45-m m nylon filters prior to batch contact testing to remove small amounts of entrained solids. The ⁹⁹Tc concentration (0.267 m Ci/mL) in the filtered AZ-102 sample was higher than the AN-103 sample. The NO₃^- and free OH^- concentrations of the filtrate were 0.27 M and 0.11 M, respectively. Tank AN-102 waste supernate contained relatively high amounts of soluble strontium and transuranics due to the presence of organic complexants (total organic carbon: ~1.9 M).

*Molarity is reported for total moles of carbon since the speciation of the carbon-containing species is unknown.

Prior to batch contact tests, the AN-102 sample was subjected to ⁹⁰Sr and transuranics removal by precipitation. The sample was then filtered by cross-flow filtration and diluted to approximately 5.98 M Na⁺ with 0.01 M NaOH filter wash solutions. The NO₃^-, free OH^-, and ⁹⁹Tc concentrations in the diluted filtrate were 1.52 M, 1.25 M, and 0.081 m Ci/mL, respectively.

The ion exchange resin used for technetium removal from Hanford tank solutions was SuperLig^Ò 639 (batch # 981015DHC720011) supplied by IBC Advanced Technologies, American Fork, Utah. This resin is a polystyrene based material, which contains a proprietary ligand that selectively removes pertechnetate from high ionic-strength alkaline solutions. It is believed that the resin adsorbs the pertechnetate as an ion pair in its sodium and potassium forms (i.e. NaTcO₄ and KTcO₄) from the Hanford waste solutions. Similar removal of pertechnetate salts by solvent extraction from alkaline solution using crown ethers has been reported [12,13]. The SuperLig^Ò 639 resin was supplied as a 20-70 mesh size range containing no adsorbed species and was used "as received" without pretreatment. The sieve analysis of the resin is shown in Table 2.

The apparatus used to perform multiple batch contact experiments included an analytical balance (Mettler Toledo, Model AE200) accurate to ± 0.001 g, polyethylene bottles, a Maxi-Mix^Ò III rotary shaker, (Type M65800) supplied by Barnstead/Thermolyne, Bubuque, IA, and nylon filter units with Nalgene, plastic filter holders obtained from Nalgene Nunc International, Rochester, NY. House-supplied vacuum and a trap assembly were used during sample filtration. The experiments were performed in a shielded cell, allowing remote handling of materials.

Sorption of pertechnetate by SuperLig^Ò 639 resin was studied using a multiple batch contact method. All tests were performed in duplicate. For each test, a known volume of the waste solution (~18 mL) was added into a polyethylene bottle with a known quantity of ion exchange resin (~0.18 g). The solution volume to resin mass ratio (phase ratio) was typically ~100:1. The bottles containing the waste solution and the resin were immediately placed on the Maxi-Mix shaker and gently shaken for 24 ± 1 hours at ambient temperature (26 ± 1°C). Control samples (~18 mL of waste solution) were treated in the same way as duplicate test samples without the addition of resin. The concentration of technetium in control samples was used as the starting initial concentration in the batch contact experiments.

After the contact period, the resin was separated from the sample solution by filtration with 0.45-m m nylon filters. A sub-sample of the filtrate was diluted (10:1) with de-ionized water to reduce the radiation dose rate when the sample was transferred from the hot cell to the analytical laboratories. All dilutions and measurements were performed based on mass and corrected for the density of the solution to ensure accuracy. The concentration of ⁹⁹Tc was determined before and after contact using inductively coupled mass spectrometry (ICP-MS) following a 100x dilution in 0.5 M nitric acid.

Following the initial batch contact tests described above, two sequential re-contacts were performed on the filtrates collected from prior tests. In the first re-contact test, a known volume (~12 mL) of the filtrate from the initial test was re-contacted with fresh SuperLig^Ò 639 resin (~0.12 g). The fresh resin and the filtrate were gently shaken for 24 ± 1 hours. After equilibration, the resin was separated from the solution and filtered with a 0.45-m m nylon filter. A sub-sample of the first re-contact filtrate was removed from the cell and analyzed by ICP-MS to determine the concentration of ⁹⁹Tc. The second re-contact was performed by gently shaking a known volume (~10 mL) of the first re-contact filtrate with fresh SuperLig^Ò 639 resin (~0.1 g) for 24 ± 1 hours. After equilibration, the resin was separated again from the solution by filtration through a 0.45-m m nylon filter. Sub-samples of the filtrate from the second re-contact tests were submitted for analysis as described above.

Following the two sequential batch contact tests, non-radioactive ⁹⁹Tc (as NaTcO₄) was spiked into known volumes of the AN-103 and AN-102 waste solutions. This was done to increase the concentrations of ⁹⁹Tc in the waste solutions by a factor of ~10 relative to the initial concentrations in the "as-received" samples. The AZ-102 sample was not spiked since the "as-received" sample already contained a high ⁹⁹Tc concentration. The spiked solutions were stirred periodically for 4 hours in the cell to promote mixing before contacting with the resin. The spiked waste solutions (~10 mL) were gently shaken with fresh SuperLig^Ò 639 resin (~0.1 g) for 24 ± 1 hours. Control spiked sample were also treated in identical steps as the duplicate spiked waste samples. After the equilibration period, the solutions were filtered with 0.45-m m nylon filters. Sub-samples of the filtrates were removed from the shielded cell and analyzed for ⁹⁹Tc by ICP-MS.

Mixtures of Hanford tank waste samples were prepared for batch distribution coefficient measurement. Small aliquots (3.0 ± 0.1 mL) of AN-102 and AZ-102 were each added into separate 30 ± 1 mL samples of AN-103 in 60-mL polyethylene bottles. A small aliquot (3.0 ± 0.1 mL) of AN-102 was also added to 30 ± 1 mL of AZ-102 sample in a 60-mL polyethylene bottle. Upon addition, the solutions were mixed by gently shaking the bottles at 15-minute intervals for 4 hours. Four hours after initial mixing of the solutions, a 24-hour batch contact experiment was initiated to measure TcO₄^- batch distribution coefficients. Twenty-four hours after initial solution mixing, the solution mixtures were visually examined for foaming, gas bubbling, color change, and visible solids formation. Two weeks after initial mixing of the solutions, a second 24-hour batch distribution coefficient measurement was conducted on each solution. In each test, a known volume of solution mixture (~5 ± 0.1 mL) was added into a polyethylene bottle with a known quantity (~0.05 ± 0.005 g) of SuperLig^Ò 639 ion exchange resin to maintain a solution-to-resin ratio of ~100:1 for most tests. The bottles containing the solution and the resin were placed on the orbital shaker and were gently shaken for 24 ± 1 hours at cell ambient temperature (26 ± 1°C). The tests were conducted in duplicate using the same method as previously described. Sub-samples of the filtrate (test and control samples) were analyzed for ⁹⁹Tc by ICP-MS.

A batch contact test was conducted on a portion of the AZ-102 sample, which had been treated for cesium removal using SuperLig 644 resin columns. The test was performed in duplicate by adding 5.0 mL of decontaminated AZ-102 solution into 0.05 ± 0.005 g of SuperLig^Ò 639 resin. After 24-h equilibration, the solutions were filtered with 0.45-m m nylon filters. Sub-samples of the filtrates were removed from the shielded cell and analyzed for ⁹⁹Tc by ICP-MS. Control samples of the decontaminated AZ-102 were also treated in identical steps as the duplicate test samples.

Pertechnetate batch distribution coefficients (K_d's) and percent removal (% R) were calculated using the following equations:

Where, C_i is the initial concentration of pertechnetate ion in control waste samples, C_f is the pertechnetate concentration in test waste samples at equilibrium, V is the volume of test or control waste samples, M is the mass of the "as-received" resin, and F is defined as the mass of dried sample divided by the mass of "as-received" sample. An F-value of 0.987 was determined for SuperLig^Ò 639 (batch # 981015DHC720011).

Multiple batch contact experiments were utilized to determine the amounts of unextractable ⁹⁹Tc (presumed to be technetium species other than pertechnetate) in each waste sample. Based on these experiments, the AN-103, AZ-102, and AN-102 samples are considered to contain 1.6, <0.1, and 63% nonpertechnetate species. Corrections were made to the ⁹⁹Tc ICP-MS data to adjust the total technetium concentrations to pertechnetate concentrations for each control and batch contact sample. The corrections involved simple subtraction of the amounts of non-pertechnetate determined for each waste sample from the measured total ⁹⁹Tc values. This method was utilized to determine the C_i and C_f values used in equations 1 and 2.

A multiple batch contact method was performed with SuperLig^Ò 639 resin to determine the distribution coefficients (K_d’s) for pertechnetate ion with actual Hanford tank solutions (Table 3). A liquid-to-solid ratio of ~100 was used in the batch contact tests and an equilibration time of 24 ± 1 hours was employed. The calculated distribution coefficients are provided in Table 3 for each batch contact test. The equilibrium TcO₄^- concentrations, K_d, and resin loading values shown are averages of the results from duplicate experiments unless otherwise indicated. Pertechnetate % removals calculated for the initial contacts were approximately 87, 91, and 89 % for the AN-103, AZ-102, and AN-102 samples, respectively. For the AN-102 sample, the % removal for total technetium for the initial contact was <30%, due to the fact that ~60% of the ⁹⁹Tc was not TcO₄^-. As seen in Table 1, the K_d values do not directly correspond to resin loadings. For example, the highest overall K_d values were observed for the AZ-102 sample (Figure 1), although resin loading was actually lower for this sample than for the AN-103 or AN-102 samples.

Pertechnetate adsorption isotherms are shown for each waste sample in Figure 2. Resin loading is plotted versus the equilibrium pertechnetate to nitrate ratios in order to normalize the impact of nitrate, which is considered to be the primary competitor with pertechnetate for SuperLig^Ò 639 adsorption sites in Hanford waste solutions. As would be expected, pertechnetate loading increases with increasing equilibrium pertechnetate to nitrate ratios, due to the decreased competition of nitrate. The resin performance based on pertechnetate loading in the various waste samples decreased in the following order: AN-103, AN-102, and AZ-102.

*Value calculated from a single batch contact experiment rather than the average of duplicate experimental results.
NA = not applicable; K_d and loading values were not calculated for this experiment because the data
indicated that no technetium was adsorbed to the resin.

The pertechnetate loading curves for each sample are self-consistent and follow a single curve. The slopes of the loading curves for all three waste samples are quite similar but the curves are slightly offset relative to each other.

The differences between the loading curves could be attributed to changes in the pertechnetate to nitrate selectivity coefficient or to changes in the effective resin capacity between waste samples. The lower loadings observed for the AZ-102 sample indicates that resin performance varies directly with the ionic strength of the solution. This is consistent with the observation that pertechnetate elutes rapidly from SuperLig^® 639 resin in deionized water [14]. The resin vendor claims that after elution the resin adsorption sites are not occupied by any adsorbed species, i.e. both sodium nitrate and sodium pertechnetate are removed from the resin at zero ionic strength. These observations are more consistent with the theory that the effective resin capacity, rather than the selectivity, varies with ionic strength.

The ultimate Hanford waste processing strategies may involve blending of various tank waste samples or processing solutions (such as waste tank heels) prior to ion exchange treatment. One of the postulated results of mixing these samples is that pertechnetate ion may be converted to a non- pertechnetate (presumably reduced) form of technetium, thereby decreasing the efficiency of the technetium removal columns.pertechnetate (presumably reduced) form of technetium, thereby decreasing the efficiency of the technetium removal columns. A simple approach to evaluating these effects involves measuring the distribution coefficient values of pertechnetate in waste solution mixtures. Distribution coefficients were determined for 10:1 volume mixtures of the AN-103 sample with the AZ-102 sample and with the AN-102 sample. Additional K_d tests were conducted on a 10:1 volume mixture of AZ-102 with AN-103. Twenty-hour batch contact tests were conducted on the mixtures after 4-hour and 14-day mixing times. Mixing the tank samples did not result in the formation of observable solids or adverse chemical reactions (such as foaming, gas release, or color change) over the 14-day storage period. The blended waste compositions were assumed to be the simple volume-weighted averages of the separate waste samples and no analyses were performed on the mixtures to confirm that unexpected compositional changes had not occurred.

An alternative mechanism for the conversion of pertechnetate to other technetium species might involve the introduction of organic compounds into the waste stream by processes in a waste treatment plant. One such source of organic species is SuperLig^® 644 degradation products from columns used for cesium removal. The impact of these organics was evaluated by determining the K_d values of the AZ-102 sample after cesium removal with SuperLig^® 644 columns.

Table 3 compares the TcO₄^- K_d’s for Hanford mixed waste solutions to those observed for the pure waste solutions. Quantitative prediction of the effects of mixing the waste solutions on the K_d values is not a straight forward process and depends on several factors such as the impact of changes in ionic strength, overall solution composition, pertechnetate concentration, and pertechnetate to total ⁹⁹Tc ratios. The K_d values for the pure waste samples are included in Table 3 only to provide a basis for evaluating whether shifts observed in the K_d values for the mixed samples relative to the pure samples are in the expected directions.

Table 4. TcO₄^- K_d Values for Hanford Mixed Waste Samples

The K_d values determined after 4-hour and 14-day storage periods for solution mixtures of AN-103 with AZ-102 (10:1 v/v) were 615 and 600 mL/g, respectively. Aging the solution for two weeks did not significantly effect the distribution coefficient relative to the value obtained after a 4-hour mixing time. This indicates that there is no slow conversion of pertechnetate to other unextractable technetium species on the time scale of the experiment.

The K_d for the mixed sample was slightly higher than was observed for the pure AN-103 sample, as would be expected based on the higher K_d observed for the AZ-102 sample relative to AN-103. Conversely, the addition of a small amount of AN-102 to AN-103 resulted in a slight decrease in the measured K_d value as would be expected based on the values observed for the pure waste samples. The TcO₄^- K_d values observed for the AN-102 mixed with AN-103 after 4 hours and 14 days were 456 and 392 mL/g, respectively. This indicates that aging the solution prior to pertechnetate removal may result in a slight decrease (14%) in the resin performance. Figure 3 shows TcO₄^- equilibrium loading as a function of the equilibrium NO₃^-/TcO₄^- molar ratio for the mixed samples predominantly made from AN-103 with AZ-102 or AN-102. As shown, the equilibrium loading of the mixed samples are very consistent with the pure AN-103 sample and the effect of the contact period (4 h vs. 14 days) appears to be minor.

For the mixture of AZ-102 and AN-102, the TcO₄^- K_d values were 592 and 945 mL/g for the 4-hour and 14-day old solutions. This drastic improvement in resin performance after aging the solution indicates that process hold times may be required to achieve acceptable technetium removal from AZ-102/AN-102 mixtures such as this. Based solely on the K_d values for the pure samples, the K_d of the mixture would be expected to decrease relative to the value observed for the AZ-102 sample (886 mL/g). The increase in the K_d of the 14-day sample indicates that mixing of AZ-102 and AN-102 samples may result in the conversion of non-pertechnetate species in the AN-102 waste to TcO₄^-. This observation could have significant process impacts. The equilibrium loading of the mixed samples prepared from AZ-102 with AN-102 and pure AZ-102 are plotted in Figure 4 as function of the equilibrium NO₃^-/TcO₄^- molar ratio. The loadings of the mixed samples (after 4 h and 14 days) are consistent with the pure as well as cesium decontaminated AZ-102 samples.

A molecular recognition product, referred to as SuperLig^® 639 was evaluated for technetium removal from Hanford tank wastes. Multiple batch contact tests were performed on the SuperLig^® 639 resin to determine the sorption characteristics of technetium as TcO₄^- anion from three Hanford tank waste solutions and their mixtures. The resin was found to be very selective for TcO₄^- removal from a wide range of tank compositions. The pertechnetate batch distribution coefficients for three Hanford tank solutions (AN-103, AZ-102, and AN-102) were significantly higher than a minimum required for Hanford waste pretreatment flowsheet to achieve acceptable decontamination factors for a single pass through the ion exchange columns. There was no indication of adverse effects from mixing of primary tank waste solutions (AN-103 and AZ-102) to process tank heels (AN-102).

This work was conducted at the Savannah River Technology Center in Aiken, SC, which is operated for the U. S. Department of energy by Westinghouse Savannah River Company under contract DE-AC09-96SR18500. The Hanford River Protection Project Office supported this work. We are very grateful to Karen N. Palmer, Yvonne J. Simpkins, Francis Wakefield, and Myra Bussey for their assistance in the completion of the experimental work. We also thank Major Thompson for his contributions toward the experimental design and review of this work.

1. K. Wildung, M. McFadden, and T. R. Garland, J. Environ. Qual. 8, 156 (1979).
2. A. J. Marrus, T. M. Gillian, and L. R. Dole, Review of EPA, DOE, and NRC Regulations on Establishing Solid Waste Performance Criteria. Report ORNL/TM-9322, U. S. Department of Energy, Oak Ridge, Tennessee, 1988.
3. S. F. Marsh, Z.V. Svitra, S. M. Bowen, Distributions of 14 Elements on 63 Absorbers from Three Simulant Solutions (Acid-Dissolved Sludge, Acidified Supernate, and Alkaline Supernate) for Hanford HLW Tank 102-SY. Report LA-12654, Los Alamos National Laboratory, Los Alamos, New Mexico, 1994a.
4. S. F. Marsh, Z.V. Svitra, S. M. Bowen, Distributions of 15 Elements on 58 Absorbers from Simulated Hanford Double-hell Slurry Feed (DSSF). Report LA-12863, Los Alamos National Laboratory, Los Alamos, New Mexico, 1994b.
5. S. F. Marsh, Z.V. Svitra, S. M. Bowen, Distributions of 12 Elements on 64 Absorbers from Simulated Hanford Neutralized Current Acid Waste (NCWA). Report LA-12889, Los Alamos National Laboratory, Los Alamos, New Mexico, 1994c.
6. S. F. Marsh, Z.V. Svitra, S. M. Bowen, Effects of Aqueous-Soluble Organic Compounds on the Removal of Selected Radionuclides from High-Level Waste. Report LA-12863, Los Alamos National Laboratory, Los Alamos, New Mexico, 1995.
7. N. C. Schroeder, S. D. Radzinski, K. R. Ashely, J. R. Ball, F. Stanmore, and G. Whitener, Technetium Partitioning for the Hanford Tank Waste Remediation System: Sorption of Technetium from DSSF and DSSF-7 Waste Simulant Using Reilex-HPQ Resin. Report LA-UR-95-40, Los Alamos National Laboratory, Los Alamos, New Mexico, 1995a.
8. D. L. Blanchard, G. N. Brown, S. D. Contradson, S. K. Fadeff, G. R. Golcar, N. J. Hess, G. S. Klinger, D. E. Kurath, Technetium in Alkaline, High-Salt, Radioactive Tank Waste Supernate: Prelimenary Characterization and Removal. Report PNNL-11386, Pacific Northwest National Laboratory, Richland, Washington, 1997.
9. K. R. Ashley, G. D. Whitener, J. R. Ball, S. D. Radzinski, and N. C. Schroeder, Solvent Extr. Ion Exch. 16, 843 (1998).
10. K. R. Ashley, G. D. Whitener, N. C. Schroeder, J. R. Ball, and S. D. Radzinski, Solvent Extr. Ion Exch. 17, 1543 (1999).
11. D. J. McCabe, N. M. Hassan, W. D. King, J. L. Steimke, M. A. Norato, L. L. Hamm, and L. N. Oji, Comprehensive Scale Testing of the Ion Exchange Removal of Cesium and Technetium from Hanford Tank Wastes, WM01 proceedings
12. M. G. Jalhoom, J. Radioanal. Nucl. Chem. 104, 131 (1986).
13. G. V. Korpusov, Y. S. Krylov, S. I. Tsalon, Sov. Radiochem. 26, 390 (1984)
14. S. Seifert, R. Muenze, P. Leibnitz, G. Reck, and J. Stach, Inorg. Chim. Acta, 193, 167 (1992).

Figure 1. Pertechnetate K_d Values for Hanford Tank Waste Supernates

Figure 3. Pertechnetate Equilibrium Loading for AN-103 Mixed with AZ-102 or AN-102