F. F. Fondeur, T. Hang, M. P. Bussey, W. R. Wilmarth,
D. D. Walker, and S. D. Fink
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 study investigated the effect of salt solution chemistry on cesium loading of IONSIV^® IE-911. The study focused on salt solutions with different levels of carbonate, oxalate and peroxide.

• Cesium loading on IE-911 increased with carbonate content in the simulated salt solution. Over the range of concentrations expected in Savannah River Site (SRS) waste, the cesium loading increases by 40%. The increase likely results from a shift in sodium activity in the solution for increasing carbonate concentration and agrees with expectation based on the ZAM equilibrium model.
• Removing oxalate from simulated "average" salt solution had no effect on cesium loading.
• Peroxide at 0.1 M decreased the cesium loading on IE-911 by 9%. The authors estimate the expected peroxide concentration in average SRS waste as 2.6 x 10^-6 M. At this concentration level and using linear extrapolation of this data, one expects no peroxide effect on cesium loading.

Keywords: Oxalate, Carbonate, peroxide, Cesium, distribution, coefficient, ion exchange. Salt Waste Processing Facility

The Savannah River Site (SRS) continues to examine three processes for the removal of radiocesium from high-level waste. One option involves the use of crystalline silicotitanate (CST) as a non-elutable ion exchange medium. The process uses CST in its engineered form - IONSIVÒ IE-911 made by UOP, LLC. - in a column to contact the liquid waste. Cesium exchanges with sodium ions residing inside the CST particles. The design disposes of the cesium-loaded CST by vitrification within the Defense Waste Processing Facility.

The site produces and stores a variety of liquid wastes with significant variation in composition. In addition, the highly caustic waste constantly absorbs carbon dioxide from the atmosphere. Therefore, the ion exchange material, IONSIVÒ IE-911, must remove cesium over a wide range of waste chemistries. The chemical elements present in the waste not previously tested for their effects on IONSIVÒ IE-911 performance – at these ranges -- include carbonate, oxalate and the radiolysis byproduct peroxide. The oxalate comes chiefly from previous additions of oxalic acid in various cleaning operations. This study examined the rate of cesium removal and capacity of IONSIVÒ IE-911 in simulated wastes containing carbonate, oxalate and peroxide.

The data from this study will be stored in notebook # WSRC-NB-2000-00203. This study originates from a R. A. Jacobs technical task request. Testing in this work is covered under D. D. Walker Task Plan.

Before conducting the cesium removal tests, personnel determined the approximate solubility of sodium oxalate and sodium carbonate in waste compositions of interest. Personnel prepared 1 liter of "average" salt solution with different amounts of sodium carbonate (corresponding to 0.5, 0.7 and 1.0 molar concentration). See Table 1 for the composition of these solutions. When adding sodium carbonate, we deducted sodium nitrate to maintain constant sodium molarity. During preparation, personnel maintained the concentration of sodium as 5.6 molar regardless of carbonate content; consequently, the ionic strength increased with increasing carbonate concentration. Researchers mixed the solution for 4 hours, filtered through a 0.02-micron nylon filter, and analyzed a portion of the solution for sodium, potassium, cesium and all the anions. Personnel followed a similar procedure to make solutions with sodium oxalate concentrations of 0.01, 0.001 and 0.0001M.

Personnel placed 5 grams of IE-911 in a glass column suspending the sorbent on a #2 mesh filter. They passed distilled water in an up-flow motion through the sorbent bed at 4 cm/min until all the fines suspended above the bed disappeared. They then passed a 2 M sodium hydroxide solution in a down-flow motion at 4 cm/min until the equivalent of 40 bed volume flowed through the column. Personnel then passed distilled water again in a down-flow motion at 4 cm/min until the pH of the liquid exiting the column fell within the range of 9 to 10. At this point, personnel permitted the bed to drain, and dried the sorbent to constant weight in a dessicator.

Loading test utilized the "average" simulated waste solution (see Table 1).

Additional tests featured simulated wastes with 0 and 0.7 M sodium carbonate in "average" salt solution at constant sodium molarity (5.6 M). The ionic strength of the solutions varied with the carbonate content as indicated in Table 2.

Personnel spiked solutions with 5 m Ci/L (¹³⁷Cs). Personnel placed 0.100 grams of pre-treated IE-911 in 20 mL of solution. The slurry shook at 150 rpm for a finite time (24, 48, 72, 120, 168 and 216 hours) in an orbital shaker. Personnel maintained the shaker temperature to within ± 1.3 ° C of 25 ° C. At the end of the shaking test, researchers filtered the slurry using a 0.02-micron nylon filter and analyzed the filtrate with a gamma spectrometer. Testing with IE-910 materials completed after 5 days with a single sample (5 days) analyzed for ¹³⁷Cs activity.

The gamma spectrometer response proved linear over the ¹³⁷Cs concentration range studied (see Figure 1). Personnel determined the Limit Of Detection (LOD) for the gamma spectrometer as 2.00 x 10^-6 M or 266 ppb cesium. In this work the cesium concentration always exceeded 266 ppb.

Figure 2 compares the performance of this spectrometer with a similar unit in the Analytical Development Section.

When determining cesium removal efficiency, the authors calculated the ratio of gamma counts for the initial (before contacting IE-911) and final solution. The following formula thus provide the cesium K_d.

This K_d value provides a scale to gage the effect of carbonate, oxalate and peroxide on cesium loading. Similarly, one can determine the amount of cesium loaded on IE-911 per gram of IE-911 using the following formula.

This quantity, q (mg Cs / g CST), proves less susceptible to errors from low gamma counts than the expression for K_d. These two quantities, q and K_d, provide a means to evaluate the effects of solution chemistry on cesium loading.

Carbonate solubility in "average" salt solution exceeded 1.0 molar. This finding seems surprising given the high ionic strength of the solution. The solubility of Na₂CO₃ is reported at 9.32 wt. % in water at 25 ° C. Thus, the ionic strength of these solutions (6.77 to 7.69 M) does not appreciably affect carbonate solubility.

Oxalate proved soluble to 0.002 M in these tests but well below that previously reported by Wiley and Kilpatrick (0.015 M at ionic strength of 6.55 M). The discrepancy strongly suggests our solution did not reach equilibrium.

Figure 3 shows the effect of carbonate content in "average" salt solution on cesium K_d values. Higher carbonate content resulted in higher cesium K_d values for IE-911. Personnel performed similar experiments (except all the experiments stopped after 5 days) with samples of IE-910 to test whether this observation proves intrinsic to CST or whether the binder used in the IE-911 causes this behavior.

Figure 4 provides the results in bar graph style. The figure affirms the observation of increasing cesium K_d values with increasing carbonate content. An F-test statistical evaluation of 12 samples – six treated in "average" salt solution with 0.16 M carbonate and six treated in "average" salt solution with 0.7 M carbonate – indicate cesium K_d values increased in proportion with the carbonate content of the salt solution.

One reaches the same conclusion when the F-test analysis of another six samples treated in "average" simulant with no carbonate proves statistically different from the "average" simulant test. The F-test value for the 0.7 M CO₃^2- simulant exceeds the value of 4.96 (z-value for 12 samples and two solutions). This calculation indicates the cesium Kd value in simulant with 0.7 M CO₃^2- differs from the corresponding value for simulant with 0.16 M CO₃^2-. One reaches a similar conclusion when comparing the F-value of the simulant with no CO₃^2- (F=26.5) to the z-value (4.96).

The lead author also ran the ZAM model to test if activity coefficient changes brought about by introducing carbonates into the salt solution can replicate the experimental observation. Table 3 lists the results from the model. The predicted cesium K_d values increased with the carbonate content in the salt solution. A change in activity coefficient may explain the positive correlation between carbonate content and cesium K_d values.

Finally, personnel analyzed the potassium, sodium and cesium content of the salt solutions with carbonate. The results indicated no significant variation in the sodium, potassium and cesium content of the salt solution as a function of carbonate concentration. This result indicates the only possible explanation for the increase in cesium K_d values involves an increase in the loading of ions on IE-911. When IE-911 comes in contact with high ionic strength solutions more ions move into the IE-911. Especially those ions which more easily shed waters of hydration. In this case cesium and potassium shed water more readily than from sodium and thus move into the IE-911. Correspondingly sodium and its associated water in the pore move out into solution to lower ion-ion interactions and hydrate the ionic solution. The authors do not imply any special sequestration of sodium or potassium by the carbonate ions.

Figure 5 depicts the effect of oxalate on cesium removal. The figure indicates no oxalate effect on cesium capacity or rate of loading. The early difference seen in cesium loading reflects experimental variance. Personnel obtained a further verification that oxalate has no effect by performing similar tests with IE-910 IE-910. Looking at Figure 6, one concludes no oxalate effect exists on cesium loading since the Kd values are not statistically different. Likewise, results from ZAM modeling predict no effect with cesium loading (2106 versus 2260 mL/g). In fact increasing oxalate concentration up to 0.1 M in the ZAM model had only a very small effect on cesium loading (data not shown).

The Effect of Peroxide

Personnel carried out simultaneously two sets of experiments. In one of the two experiments (control) researchers injected "average" salt solution containing 0.005 g/mL of IE-911 every five hours with 100 m L of distilled water. In the other set, personnel injected "average" salt solution containing 0.005 g/mL IE-911 every five hours with 100 m L of 50 wt % hydrogen peroxide solution. The peroxide concentration, as determined by permanganate titration, equaled 0.13 M immediately after injection. The peroxide concentration decreased to 0.0034 M five hours later. Both sets of experiments occurred on the same shaker. Figure 8 presents the results. The data points in Figure 8 represent the average of duplicates. A look at the IE-911 data indicates a lower cesium loading in the in the presence of peroxide (see Figure 8). At this level of peroxide, the cesium Kd Value decrease by approximately 25 %.

The authors conducted the same type of experiment with IE-910. Figure 9 displays that data. An inspection of the data indicates a marginally lower cesium loading in the presence of peroxide.

The tests used a conservatively high concentration of peroxide. Subsequently the authors estimated the expected peroxide level in a cesium loaded plant column as about 2.6 x10^-6 M based upon radiolysis measurements in water at half the dose rate. The authors obtained the estimated value at 1 Mrad/h by doubling the value of peroxide production at 0.5 Mrad/h. At this level of peroxide, linear interpolation of the data in Figure 8 suggests no effect from the peroxide.

This study investigated the effect of changes in the solution chemistry on cesium loading. Experiments examined the influence of varying the concentration of carbonate, oxalate and peroxide on cesium removal by either IONSIV^® IE-910 or IE-911. The study found the amount of cesium loading on IE-911 varies proportionally with the amount of carbonate in solution. This correlation results from an increase in the ionic strength of the solution with carbonate content. Higher ionic strength decreases the water activity in the bulk solution. More easily dehydrated cesium and potassium ions become load on to the resin, displacing sodium ions. The presence or absence of oxalate in solution had no effect on cesium loading. Peroxide at a 0.13 M concentration, lowered cesium loading on IE-911. The expected peroxide concentration in the column (2.6 x 10^-6 M) is not expected to significantly affect cesium loading.

1. R. A. Jacobs, HLW Technical Task Request, "Post-precipitation in Simulants and Capacity/Kinetics of IE-911," HLW-SDT-TTR-99-37.1, December 20, 1999.
2. D. D. Walker, W. R. Wilmarth, F. F. Fondeur, and T. Hang, "Task Technical and Quality Assurance Plan for Non-Elutable Ion Exchange Process Waste Stability and IONSIVÒ IE-911 Performance Tests," WSRC-RP-99-01079.
3. D. D. Walker, "Preparation of Simulated Waste Solutions," WSRC-TR-99-00116, Rev. 0, March 15, 1999.
4. David R. Lide, "CRC Handbook of Chemistry and Physics," 77^th Edition, CRC Press, 1996-1997.
5. J. R. Wiley, "Sodium Oxalate Solubility in Simulated SRP Waste Solutions," DPST-78-480, August 1979.
6. L. L. Kilpatrick, "Solubility of Sodium Oxalate and Sodium Tetraphenylborate in DWPF Supernate," DPST-84-314, February 1984.
7. J. T. McClave and F. H. Dietrich, II, "Statistics," Second Edition, Dellen Publishing Company, 1982.
8. Z. Zheng, R. G. Anthony, and J. E. Miller, "Modeling Multicomponent Ion Exchange Equilibrium Utilizing Hydrous Crystalline Silicotitanates by a Multiple Interactive Ion Exchange Site Model", Industrial Engineering Chemistry Research, March 1995.
9. R. D. Whitley and N. H. L. Wang, "User Manual-VERSE Simulation for Liquid Phase Adsorption and Chromatography Processes," Purdue University, School of Chemical Engineering, September 1998.
10. A. Chatterjee, J. L. Magee, and S. K. Dey, "The Role of Homogeneous Reactions in the Radiolysis of Water," Radiation Research, Vol. 96, PP. 1-19 (1983).