L. N. Oji and D. T. Hobbs
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 laboratory experiments performed to examine the feasibility and rate of isotopic dilution of uranium-235 in Tank 43H supernatant liquid. The isotopic dilution was accomplished by adding an alkaline depleted uranium solution to simulated or actual Tank 43H waste solutions having enriched U-235 compositions. The laboratory testing results indicated the following:

• We observed rapid dilution of the U-235 enrichment in the supernatant liquid to less than 0.5 percent upon addition of an alkaline solution containing depleted uranium.
• The depleted uranium delivery mode and addition rate were not important parameters affecting isotopic dilution in the supernatant liquid.
• Mixing is an important parameter affecting isotopic dilution in the supernatant liquid.
• Under appropriate mixing conditions, isotopic dilution of uranium-235 is faster than precipitation of depleted uranium.
• With effective mixing, isotopic dilution to an acceptable level occurred in minutes.
• Where there is limited or no effective means of mixing the depleted uranium solution with the supernatant liquid containing enriched uranium, instantaneous bulk addition of the depleted uranium appears more effective in ensuring appreciable isotopic dilution than a slow measured delivery of depleted uranium.
• Slower addition of the depleted uranium solution is beneficial for isotopic dilution only when it is accompanied by effective mixing.
• Precipitated solids formed upon addition of depleted uranium solution are micron size particles, which are easily suspended in solution with minimal stirring.

Analyses of Tank 43H supernatant samples indicates that the supernate has an average total uranium concentration of 15.0 ± 2.9 mg/L and an average uranium-235 enrichment of 3.12 ± 0.63 weight percent. Because of this high uranium-235 enrichment in Tank 43H, which serves as a feed tank to the 2H-evaporator plans are to add depleted uranium (DU) as a uranyl carbonate solution to Tank 43H to reduce the U-235 enrichment in the supernatant liquid. Reducing the U-235 enrichment below that of natural uranium abundance ensures that the 2H-Evaporator can be operated safely without risk of criticality. The target uranium-235 enrichment for the Tank 43H supernate is less than 0.96 weight percent.

To ensure adequate mixing of the uranyl carbonate solution with the tank supernate, plans are to install a Flygt mixer, positioned about 3.3 meters from the sludge and pointed up at an angle. The uranyl carbonate solution will be introduced at the Flygt mixer.

This report documents experimental results conducted at the Savannah River Technology Center (SRTC) to examine the feasibility, extent and rate of isotopic dilution of uranium-235 resulting in the addition of an alkaline depleted uranium solution to Tank 43H supernate simulant and to the real waste supernatant liquid currently stored in Tank 43H. The sludge at the bottom of Tank 43H storage tank is not taken into consideration in the simulation and actual waste tests.

This work complies with the work scope defined in the following plans: D.T Hobbs and L. N. Oji, " Task Technical and Quality Assurance Plan for Testing Methods to Reduce ²³⁵Uranium Enrichment in Tank 43H supernatant Liquid," WSRC-RP-2000-0097, Rev.0, Sept. 20,2000 and Technical Task Request HLE-TTR- 2000-063 from B.L. Lewis titled "Development of process to reduce the Uranium Enrichment in Tank 43H," August 17, 2000.

Keywords: Uranium-235, Uranium enrichment, Fissile material, 2H-Evaporator

2.10 Approach

The test plan, summarized in section 2.3.0, features the use of an alkaline DU solution to obtain the desired isotopic dilution of U-235 in both the enriched Tank 43H simulated supernate and in the real waste. In these tests, 4 ml DU solution (10 g U/L), an amount needed to reduce the U-235 enrichment well below 0.96 weight percent, was added to 100 ml of the simulant or real waste under various delivery and mixing conditions. An inductively coupled plasma- mass spectroscopic technique was used to quantitatively measure the uranium isotopic distribution.⁶ Efficiency of the supernate isotopic dilution process was based on the differences, if any, in the magnitude of experimental isotopic distribution data obtained versus the expected values. How good the isotopic dilution process worked experimentally was also judged on changes in the equilibrium uranium-235-to-uranium-238 ratio with mixing time and visual observations related to the behavior of uranium precipitates formed.

Test conditions were defined to monitor for the effects of DU delivery rate, mixing rate, and variation of U-235/U-238 ratio with addition of DU on the efficiency of isotopic dilution of the supernate. Upon addition of the large quantity of DU the following approximation (Other uranium isotopes are present in low concentrations prior to the addition of DU and subsequently fall below detection levels for the ICP-MS analytical method upon the addition of DU and preparation of the samples for the ICP-MS analysis.) holds:

U-235/U-238 » U-235/U-total.

The real waste work was performed in the SRTC shielded cells and the simulant work in a radioactive hood. The temperature in the radioactive hood and in the shielded cells were, respectively, 22 ± 0.5 and 22 ± 1° C. All results in this report are given in terms of U-235/U-238.

2.20 Materials and Solutions

The composition of the Tank 43H simulant used in this study featured the most recently reported values from dip samples taken from Tank 43H (see Table 1). Appendix A provides complete details of the Tank 43H simulant preparation. The target U-235 enrichment for the simulant of 4.37 percent is based on the average U-235 enrichment from several samples measurements plus two times the standard deviation of this data set. Similarly, the target total uranium concentration is based on the average total uranium concentration of 18.95 mg/l plus approximately one times the standard deviation. The density of the Tank 43H simulant is 1.21 g/ml.

The depleted uranium solution used in these tests was prepared as follows. Uranyl nitrate (UO₂ (NO₃)₂.6H₂O, 21.1 g) dissolved in approximately 5 ml of 1.5 M nitric acid was added to a one liter volumetric flask containing 500 ml of a 0.37 M sodium carbonate (Na₂CO₃.H₂O) solution . The flask was then filled to the mark with additional sodium carbonate. Analysis of this alkaline DU solution for uranium-238 concentration gave a value of 10.0 mg/ml and U-235/U-238 ratio of 0.30 ± 0.02 percent. The measured density of this DU solution was 1.04 g/ml.

Due to insufficient archived Tank 43H supernate samples, the real waste used was a composite of filtered real waste from Tank 43H and Tank 38H. Tank 38H material was chosen since this tank is the concentrate receipt tank for the 2H-Evaporator and thus contains concentrated waste originally present in Thank 43H. Based on volume measurements, 71% of the composite sample came from Tank 43H and 29% from Tank 38H. It is worth noting that the 2H-Evaporator processes radioactive waste from tanks 43H (feed tank) and Tank 38H, with Tank 38H acting as the concentrate waste receipt tank. Therefore, we assumed that the supernate from the two tanks is similar in composition.

The mixing of depleted uranium with Tank 43H simulant or real waste carried out in capped transparent 250-ml Teflon with receiving holes on each cap as shown in Figures 1 and 4. These holes held rubber tubing directly connected to a syringe pump.

Table 1. Tank 43H Simulant Composition.

2.30 Test Design

The total uranium U-238 concentration in the depleted uranium soloution used was 10 mg/l with an average U-235/U-238 ratio of 0.30 ± 0.02 percent. Similarly, the measured uranium-238 for Tank 43H simulant was 18.95 mg/l with a U-235 concnetration of 0.766 mg/l. Based on these uranium the addition of 4 ml of the DU solution to 100-ml of the simulant solution provides a theoretical U-235/U-238 ratio of 0.00469 (0.1966/41.895).

A comparison of the experimental value for U-235/U238 ratio in the filtrate with this theoretical value of 0.00469, along with the time required to attain this equilibrium value, was used as a measure of the extent of isotopic dilution for tests with simulants and real waste.

The test plan for the simulant work consisted of three principal test categories:

• Addition rate test: Designed to evaluate the effects of depleted uranium addition rate (instantaneous addition versus slow addition at two rates: 0.267 and 0.013 ml/min.) under well-mixed and poorly mixed conditions.
• Utility test: Designed to evaluate changes in U-235/U-238 ratio with increase in DU solution addition (incremental test). Here, we maintained the volume of the simulant constant at 100 ml for nine different runs and increased the volume of DU (0.5, 1, 2, 3, 4, 10, 15, 20 and 25 ml) to see if the desired U-235/U-238 ratio in the DU solution can be attained. This test was performed in a well-mixed condition and with instantaneous addition of the DU solution.
• Mixing test: The effects of mixing on the isotopic dilution experiments were evaluated by running experiments with and without effective mixing using either a magnetic stirrer or an orbital shaker, or a combination of both mixing devices.

Samples were collected at the following time intervals; 3, 9, 24, 48, 72, 96 and 120 hours after adding depleted uranium to the simulant or real waste supernate. Typically, approximately 2 ml of the test mixture were withdrawn using a 10-ml plastic syringe filter and filtered after connecting a syringe filter disc (0.45-micron membrane). In some case, an unfiltered sample was submitted for analysis (see appendix B).

The filtered solution (1.0 ml) was delivered into a sampling vial containing 1.0 ml of dilute nitric acid (1.5M). All samples were analyzed by ICP-MS for uranium isotopes.

3.1.1 Addition Rate Tests

3.1.2 Instantaneous addition of depleted uranium

In a typical test, 100 ml of the simulant was measured and delivered into a Teflon bottle containing a magnetic stirring bar. The magnetic stirrer was operated in a manner to avoid the introduction of excess air into the solution. While the simulant solution was stirring, 4 ml of the DU solution was measured and delivered into the mixing simulant within 3 seconds. Three hours after the introduction of the DU into the solution the 3-hour samples (filtered and unfiltered) were collected. Six hours later the 9-hour samples were collected. Thereafter, samples were collected at 24-hour intervals after the initiation of the test. Appendix B shows the analytical results for all samples collected and Figure 2 shows a typical isotopic dilution profile for the instantaneous addition of DU solution into the Tank 43H simulant.

The average U-235/U-238 ratio from the seven post-DU addition samples obtained as part of this test was 0.458 ± 0.024 percent. The U-235/U-238 ratio for the filtered samples collected after 3 and 120 hours, respectively were 0.472 and 0.487 percent. These results compare favorably with the expected U-235/U-238 ratio of 0.469 percent.

Figure 1. From left to right (1) A continuously stirred mixture of 100 ml of Tank 43H
simulant and 4 ml of DU solution, (2) Tank 43H simulant (U-235/U-238 = 4.32%), and (3)
A continuously stirred mixture of 100 ml of Tank 43H simulant and 15 ml of DU solution.

Figure 2. Isotopic dilution profile for instantaneous addition of 4 ml DU solution to
100 ml of Tank 43H simulant under well mixed conditions.

We ran this test in duplicate as described in section 3.1.2 above, except that the 4 ml DU was delivered into the reaction bottle at 0.267 ml/minute using a syringe pump. The syringe pump delivery orifice was coupled to the Teflon bottle via rubber tubing connected to the Teflon cap as shown in Figure 3. The complete delivery of 4 ml DU into the simulant solution took about 15 minutes. Figure 4 shows a typical isotopic dilution profile for depleted uranium delivered at 0.267 ml/minute to the enriched simulant.

The average U-235/U-238 ratio for the first test was 0.478 ± 0.025 percent. The U-235/U-238 ratio for the filtered sample collected after 3 and 120 hours of running the test, respectively are 0.440 and 0.480 percent.

The corresponding values for the duplicate test are 0.454 ± 0.026 percent, and 0.410 percent (3 h. sampling) and 0.460 percent (120 h. sampling). The U-235/U-238 ratios for the unfiltered samples at the same times (still containing all of the uranium precipitates) are 0.440 percent and 0.470 percent, respectively.

This experimental U-235/U-238 ratio of 0.478 ± 0.025 percent for the filtrate agrees favorably with the expected value of 0.469 %. Similarly, the average U-235/U-238 ratio for the unfiltered samples, 0.454 percent, is within the one sigma uncertainty value of the expected value. These results indicate that isotopic dilution equilibrium has been reached between the supernatant liquid and precipitated uranium solids.

In this test, the 4 ml DU solution was delivered into the 100 ml enriched Tank 43H simulant solution using a syringe pump calibrated to deliver 0.013 ml of the DU per minute. Hence, it took approximately 300 minutes to deliver the entire 4 ml of the DU solution into the test bottle. The test mixture was magnetically stirred during the 120- hours test period. Because of the approximately 5 hours required to deliver the DU solution into the test bottle the first sampling was carried out at the ninth hour of mixing. Figure 5 shows a typical isotopic dilution profile for tests in which the DU was delivered at 0.013 ml/minute.

The average U-235/U-238 ratio obtained is 0.488 ± 0.096 percent. The U-235/U-238 ratio for the filtrate sample collected after nine and 120 hours of running the test, respectively are 0.380 and 0.483 percent. These values are not statistically different than the theoretical value of 0.469 percent.

Figure 3. Syringe pump set up for the delivery of depleted uranium
into U-235 enriched Tank 43H simlant.

Figure 4. Isotopic dilution profile resulting from the titration of 4 ml of
DU solution into 100 ml of the Tank 4H simulant at 0.267 ml/minute
under well-mixed condition

Figure 5. Isotopic dilution profile resulting upon the addition of 4 ml of DU solution
into 100 ml of Tank 4H simulant at 0.013 ml/minute under well-mixed conditions.

This set of tests was designed to evaluate the efficiency of the isotopic dilution process by varying the amount of added DU solution and comparing the resulting U-235/U-238 ratio. At high quantities of DU solution, U-235/U-238 ratio will approach the DU enrichment value of 0.3 weight percent.

The test plan described six DU solution addition tests, (addition of 0.5, 4, 10, 15, 20 and 25 ml) of DU solution to 100 ml of the Tank 43H simulant). These test mixtures were continuously stirred throughout the test duration (120 hours). Based on the test results we conducted three additional tests featuring intermediate DU additions (i.e., 1, 2, and 3 ml of DU solutions). These additional tests provided additional data to more accurately define the U-235/U-238 ratio curve as a function of the quantity of DU solution added to the system. Due to time constraints and previous results from testing above, we only sampled at the end of the test duration for the latter three tests. Table 2 provides a summary of the results for the entire efficiency test. The values reported in Table 2 are all filtered sample data.

In Table 2, the theoretical equilibrium value for each test is provided in column 3 versus the experimental value provided in column 2. Column 4 shows the average depleted U-235/U-238 ratio of 0.3 percent. Figure 6 shows a plot of the variation in amount of DU added with changes in enrichment of the simulant solution. This Figure also shows an overlay plot of the enrichment in the depleted uranium as a reference. In general there is good agreement between the theoretical and measured values for the isotopic ratios as a function of added DU. We attribute the differences between the two values to experimental errors associated with sampling and analysis by ICP-MS.

Table 2. Utility Test Summary: Instantaneous Addition of 4 ml DU
to 100 ml of Tank 43H Simulant.

Figure 6. Variation of the amount of DU solution Added with U-235/U-238
ratio in Tank 43H Simulant3.10 Simulant Testing

Mixing tests features a new batch of Tank 43H simulant with a U-235 enrichment of 4.69 percent.

In this test, 4 ml of the DU solution was added within 3 seconds into the reaction vessel containing 100 ml of the Tank 43H simulant under well-mixed condtions. The mixture was vigorously stirred without excessive vortex for five minutes and then transferred to an orbital shaker maintained at 44 rpm for the test duration.

The average U-235/U-238 ratio obtained is 0.557 ± 0.051 percent. The U-235/U-238 ratio for the filtrate sample collected after 3 and 120 hours are 0.550 and 0.550 percent, respectively.

The mixing test in section 3.3.1 was repeated. This time the mixture was not magnetically stirred during the introduction of DU. After the addition of DU the mixture was transferred onto an orbital shaker maintained at 44 rpm for the test duration (120 h.)

The average U-235/U-238 ratio obtained is 0.477 ± 0.027 percent. The U-235/U-238 ratio for the filtrate sample collected after three and 120 hours of running the test, respectively are 0.495 and 0.461 percent. See Figure 7 for the isotopic dilution profile.

The experimental U-235/U-238 values obtained by mixing DU and enriched simulant with moderate or vigorous stirring of the mixture (instantaneous addition), respectively, 0.477 ± 0.042 and 0.557 ± 0.051 percent. These results are not statistically different from the expected U-235/U-238 ratio of 0.469 %.

In this test, the 4 ml DU solution was delivered into the 100 ml enriched Tank 43H simulant solution using a syringe pump calibrated to deliver 0.013 ml of the DU solutionper minute. Hence, it took approximately 300 minutes to deliver the entire 4 ml DU into the simulant solution. The solution was magnetically stirred just before the first drops of DU were delivered until the end of the 120 hours test period. Because of the approximately 5 hours required to deliver DU into the enriched simulant the first sampling was done only at the ninth hour of mixing.

The average U-235/U-238 ratio obtained was 0.442 ± 0.046 percent. The U-235/U-238 ratio for the filtrate sample collected after nine and 120 hours of running the test are 0.400 and 0.410 percent, respectively.

Figure 7. Isotopic Dilution Profile for an Initially Unstirred Mixture of Instantaneously
Added 4 DU Solution to Tank 43H Simulant.

In this test, the DU delivery was similar to the description in test 3.3.3 above (0.013 ml/minute), however, the solution was not magnetically stirred during the entire 300 minutes of the DU delivery. As the DU was delivered at this rate without mixing, drops of the DU solution fell to the bottom of the reaction vessel producing two distinct liquid phases (see Figures 8). At the end of the DU solution delivery, solid particles are observed in the bottom of the test bottle. The mixture was put in the shaker at 44 rpm.

Figure 9 shows the isotopic dilution profile for this test. Equilibrium U-235/U-238 ratio was not attained within the 120 hours of running the experiment as observed in other well-mixed tests. The average U-235/U-238 ratio obtained is 0.945 ± 0.377 percent. The U-235/U-238 ratio for the filtrate sample collected after nine and 120 hours of running the test, respectively are 0.470 and 0.470 percent. Values of U-235/U-238 ratio (see Figure 9) for periods between initial and final sampling varied. This significant variation, which was not observed in other experiments, may be attributed to either sampling location or real time effects.

The U-235/U-238 ratio obtained by slowly adding DU solution (0.013 ml/minute) into the Tank 43H simulant, with vigorous stirring is comparable in magnitude (0.442 ± 0.046 %) to the ratios obtained with instantaneous addition of DU under similar mixing conditions. However, the delivery of DU into the enriched simulant with no immediate stirring to ensure effective mixing, produced isotopic dilution ratios higher than the theoretical values of 0.469 percent.

Figure 8. Depleted uranium delivered into both reaction vessels at 0.013 ml/
minute. Effective stirring in vessel A and No stirring in B. Vessel B
shows two distinct phases with settled solids on the bottom.

Figure 9. Isotopic dilution profile for unstirred and poorly mixed solution
of depleted uranium solution and Tank 43H simulant.

The real waste supernatant liquid used in this part of the investigation was a composite sample from Tanks 38H and 43H. The total filtered sample volume was 350 ml with Tank 38H comprising 29% of the total waste volume. Initial uranium isotope concentration in the composite sample was analyzed in triplicate to give an average U-235/U-238 ratio of 2.67 percent. The uranium-238 concentration was 18.957 mg/L. See Table B2-1 in Appendix B2 for details.

As in Tank 43H simulant tests discussed previously, we planned to add 4 ml of the DU solution to 100 ml of the composite waste solution. Based on the analytical information in Table B2-1 and the previously reported uranium concentration and isotopes for DU solution, we calculated the total uranium and uranium isotopes upon addition of 4 ml of the DU solution to 100 ml of real waste composite solution. The calculated quantities are 41.18957 mg uranium 238 and a U-235 enrichment of 0.407 percent.

Based on the results obtained with the Tank 43H simulants, three types of tests with real waste were planned:

Slow delivery of the DU solution into real waste solution under well-mixed conditions using a magnetic stirrer and a syringe pump to deliver DU at 0.267 ml per minute (Active mixing of DU and real waste).

• Slow delivery of DU into real waste under well-mixed conditions using a magnetic stirrer and a syringe pump to deliver DU at 0.267 ml DU per minute. At the end of DU delivery, the mixture is put onto an orbital shaker set at slow shaking pace of 44 rpm (Active mixing of DU and real waste followed by gentle rocking of mixture).
• Depleted uranium is delivered into the real waste at 0.267ml DU per minute using at syringe pump, but this time the mixture is not magnetically stirred (Slow delivery of DU into real waste with poor mixing).

The real waste test was performed in SRTC Shielded Cell 10, which has a high background radiation. As a result of the high radiation environment in cell 10, the digital syringe pump after functioning well during pre-test evaluations malfunctioned after a couple of days in cell 10. Consequently, we modified the test plan to add Du solution in small increments of 500 m L every five minutes while magnetically stirring.

This modification is supported by simulant testing reported above which show that the rate of DU solution addition was not an important parameter as the mixing conditions for isotopic dilution of U-235 in the supernatant liquid.

The DU and real waste mixture was actively stirred with a magnetic stirrer through out the duration of the test.

The average U-235/U-238 ratio obtained with the real waste using this approach was 0.397 ± 0.015 percent. The U-235/U-238 ratio for the filtrate samples collected after 3 and 120 hours were 0.420 and 0.370 percent, respectively (see Figure 10 and Appendix B2). The U-235/U-238 ratio for the unfiltered samples during this same period (containing all of the uranium precipitates) is 0.390 and 0.400, respectively.

.
Figure 10. Isotopic dilution profile for composite waste with the addition of 4 ml
DU solution to 100 ml of real waste under well-mixed conditions.

The real waste and DU were vigorously mixed with a magnetic stirrer during DU delivery followed by gentle rocking of the mixture in an orbital shaker (44 rpm)for the duration of the test.

The average U-235/U-238 ratio obtained with the real waste was 0.426 ± 0.021 percent. The U-235/U-238 ratio for the filtrate samples collected after 3 and 120 hours were 0.410 and 0.430 percent, respectively. The U-235/U-238 ratio for the unfiltered samples during this same period (containing all of the uranium precipitates) was 0.410 and 0.420, respectively (see Figure 11 and Appendix B2).

Figure 11. Isotopic dilution profile for real waste with the addition of 4 ml DU solution
to 100 ml of real waste under initial well-mixed condition followed by poor mixing.

3.4.3 Depleted Uranium Addition Without Stirring (Test C)

Mixture was not magnetically stirred while delivering the DU solution into the real waste solution. After delivery of the DU the test bottle was put on an orbital shaker at 44 rpm.

The average U-235/U-238 ratio obtained with the real waste was 0.407 ± 0.014 percent. The U-235/U-238 ratio for the unfiltered samples collected after 3 and 120 hours were 0.391 and 0.421 percent, respectively. The U-235/U-238 ratio for the unfiltered samples during this same period (containing all of the uranium precipitates) was 0.391 and 0.403, respectively (Figure 12 and Appendix B2).

Figure 12. Isotopic dilution profile for composite waste with the addition of 4 ml DU to
100 ml of real waste. Instantaneous addition of DU with poor mixing.

Figure 13A and B. After the mixing and continuous stirring of depleted
uranium with Tank 43H simulant for 120 hourst he magnetic stirrer was turned off
and the solution allowed to settle under laboratory conditions.

Table 3. Summary of Experimental Results for Simulant and Real Waste Tests.

Table 3 is a summary of the isotopic dilution results for both the simulant and real waste investigations. From left to right, this table shows the experimental designs, initial U-235 enrichment in the simulant and real waste, mixing conditions, expected or theoretical U-235/U-238 ratios and observed U-235/U-238 ratios.

The addition rate tests were performed to evaluate the affect of depleted uranium addition rate on isotopic dilution under well-mixed conditions. The measured isotopic dilution ratios for the three DU solution addition rate tests were not statistically different and agreed with the expected value of 0.469 percent. In all of these tests, the isotopic dilution ratios of the mixture dropped from 4.32 percent to below 0.5 percent. Thus, we conclude that the rate of DU addition did not significantly affect isotopic dilution provided the system was well mixed.

The isotopic dilution ratios obtained after 3 hours were almost identical in magnitude to the ratios obtained at end of test (120 h). This indicates that equilibrium U-235/U-238 ratio is rapidly established within the first few minutes of adding the depleted uranium with the Tank 43H simulant and that the dilution of the U-235 is faster than depleted uranium precipitation under well mixed conditions.

The measured utility curve, variation of U-235/U-238 ratio with increase in amount of depleted uranium added, is in good agreement with the predicted curve (see Figure 6). Thus, under well mixed conditions, isotopic dilution in the supernatant proceeds smoothly upon addition of the DU solution with little of the DU precipitating before mixing and diluting the U-235 in the supernate.

In the mixing tests, the idea was to determine the effects of various mixing conditions (vigorous magnetic stirring versus orbital shaker-based rocking of the bulk mixture) on the attainment of isotopic dilution. The syringe pump based delivery of depleted uranium into a magnetically stirred U-235 enriched simulant provided the best isotopic dilution results, when compared to the expected theoretical value of 0.469 percent. On the other hand, an equivalent delivery method of depleted uranium into an unstirred enriched solution gave the worst isotopic dilution results. In this unstirred blend, the mixing was limited to rocking the mixture in an orbital shaker. The observed U-235/U-238 ratio was 0.945 ± 0.377 percent with a range of 0.47 to 1.32 percent (see Figure 9). This deviation of this value from the expected U-235/U-238 ratio 0f 0.469 is quite significant.

One other interesting observation in this particular test, rests in the fact that a similar test with depleted uranium added all at once (instantaneous addition) and with minimal mixing, too, gave a significantly different result. This time the U-235/U-238 ratio obtained, 0.477 ± 0.027 percent, is in agreement with the expected value of 0.469 percent. This was also an isotopic dilution from 4.69 percent to 0.477 ± 0.027 percent and the values were independent of the sampling location in the mixing vessel.

The only difference between this test and the previous one was in the manner the depleted uranium was added to the Tank 43H simulant; instantaneous addition versus slow addition rate (0.013 ml/minute). Both tests were not stirred, but gently rocked at 44 rpm using an orbital shaker. In the slow addition case, the bulk of the DU solution did not disperse into the Tank 43H simulant resulting in two separate liquid phases. A small amount of solids formed and settled to the bottom of the vessel. Using an orbital shaker to rock the bulk of the liquid from side to side, did not result in depleted uranium diffusion into the bulk of the Tank 43H simulated solution. This test exhibited a poorer isotopic dilution response.

This result suggests that bulk addition of depleted uranium may be more effective in ensuring appreciable isotopic dilution than slow depleted uranium addition in the absence of effective mixing.

All the real waste tests were essentially performed by instantaneous addition of the depleted uranium to the real waste. However, the method of mixing varied among the three tests. In test A, the depleted uranium and waste solution mixture was stirred continuously using a magnetic stirrer. In test B the mixture was magnetically stirred while the 4 ml of depleted uranium was added into the real waste solution at 0.5 ml every five minutes. After the addition of depleted uranium solution, the mixture was placed into an orbital shaker and rocked at 44 rpm. In test C the depleted uranium was added to the real waste at 0.5 ml every 5 minutes without any stirring and then transferred to the orbital shaker and rocked at 44 rpm for the duration of the experiment.

In all three tests, the value of U-235/U-238 ratio obtained was in excellent agreement with the expected or theoretical value of 0.407 percent. This again confirms the findings from simulant testing that the desired isotopic dilution can be affected provided the DU is added in such a manner that it disperses through out the waste solution.

One important aspect of this investigation, which did not receive sufficient attention, was the monitoring and characterization of solids formed upon the addition of the depleted uranium to the simulated or real waste solution. After the mixing of the depleted uranium with the Tank 43H simulant for 120 hours, the stirring mechanism was turned off and the solution allowed to settle at ambient laboratory conditions. Within a few minutes, one observed fine solids slowly settling to the bottom of the bottle with a clarified solution layer at the top (see Figures 13A and 13 B. These changes in clarity and settling of the particles were more pronounced after leaving the solution overnight. As a general qualitative observation, the solids were easily re-suspended in solution using a magnetic stirrer at low speeds. With the solution left overnight again, most of the solids settled to a fine "powder" on the bottom of the mixing Teflon vessel.

We have used laboratory testing to evaluate the effectiveness of diluting U-235 enrichment of simulated and real waste solutions upon addition of an alkaline solution of depleted uranium.

The experiments were designed to provide insight into the following regions of Tank 43H:

• Regions near the Flygt mixer with maximum mixing conditions,
• Regions farthest away from the Flygt mixer where there is bulk solution movement but limited mixing and
• Intermediate regions of the tank where well-mixed liquid phases are in transition to areas with limited mixing

On a laboratory-scale, we used moderate magnetic stirring without excessive vortex to simulate the first mixing condition above and an orbital shaker set at low speed (44 rpm) to simulate the last mixing condition. A transfer of a well-mixed liquid from a magnetic stirrer to an orbital shaker was used to simulate the second mixing region above.

The following conclusions are based on the observations and results obtained in the laboratory-scale investigations detailed above:

• We observed rapid dilution of the U-235 enrichment in the supernatant liquid to less than 0.5 percent upon addition of an alkaline solution containing depleted uranium.
• The depleted uranium delivery mode and addition rate were not important parameters affecting isotopic dilution in the supernatant liquid.
• Mixing is an important parameter affecting isotopic dilution in the supernatant liquid.
• Under appropriate mixing conditions, isotopic dilution of uranium-235 is faster than precipitation of depleted uranium.
• With effective mixing, isotopic dilution to an acceptable level occurred in minutes.
• Where there is limited or no effective means of mixing the depleted uranium solution with the supernatant liquid containing enriched uranium, instantaneous bulk addition of the depleted uranium appears more effective in ensuring appreciable isotopic dilution than a slow measured delivery of depleted uranium.
• Slower addition of the depleted uranium solution is beneficial for isotopic dilution only when it is accompanied by effective mixing.
• Precipitated solids formed upon addition of depleted uranium solution are micron size particles, which are easily suspended in solution with minimal stirring.

Analysis of Tank 43H samples obtained after initiating this research program, indicate an average total uranium concentration in the supernatant liquid of 7.86 ± 6.92 mg/l and a U-235 enrichment of 3.83 ± 0.83 percent. This uranium concentration is about a factor of two lower than previous analyses and that which was tested in this study. The lower uranium concentration in the Tank 43H supernatant liquid reflects a less saturated or supersaturated condition with respect to uranium compared to earlier conditions. Consequently, the current Tank 43H-supernate composition may be easier to isotopically dilute than that tested in this program since more of the depleted uranium would dissolve before precipitating.

We recommend the continued work on the following items:

• Compare energy input of magnetic stirring to that in the Flygt mixer.
• Characterize the solids formed upon the addition of the depleted uranium for particle size distribution and settling rates.

This study fulfills the activity defined in "Task Technical and Quality Assurance Plan for Testing Methods to Reduce ²³⁵Uranium Enrichment in Tank 43H supernatant Liquid," WSRC-RP-2000-0097, Rev.0, Sept. 20,2000.

Data obtained from this study reside as records in WSRC-NB-99-00200.

The authors thank Analytical Development Section personnel, in particular, B. Boyce and M. Malek for performing uranium analysis. We also thank M. S. Blume, H. L. Thacker and D. Wheeler for assistance in completing the isotopic dilution testing.

1. W. R. Wilmarth and R. A. Peterson, "Analyses of Surface and Variable Depth Samples from Tank 43H," WSRC-TR-2000-00208, Rev. 0, September 2000.
2. C. S. Boley, M. C. Thompson, W. R. Wilmarth, "Technical Basis for the 242-16H Evaporator Cleaning Flowsheet (U)," SRS Report WSRC-TR-2000-00211, Rev. 0, July 12, 2000.
3. B. Lewis, "Development of Process to Reduce the Uranium Enrichment in Tank 43," Technical Task Request HLE-TTR-2000-063, Rev. 0, August 17, 2000.
4. P. D. d’Entremont ^{, "}Enrichment and Location of Uranium Precipitates from Uranyl Carbonate Addition to Tank 43H," WSRC-TR-2001-00113, Rev. 0, March 7,2001.
5. P. D. d’Entremont, " Modeling for Uranyl Carbonate Addition Process," HLW-STE-2001-00009, January. 8,2001.
6. "Inductively Coupled Plasma – Mass Spectrometry Elemental and Isotopic Analysis for Aqueous Liquid Samples," Manual 16.1, Analytical Development Section Analytical Operating Procedure ADS-1543, Rev. 1, April 1, 1998.
7. W. R. Wilmarth, "Analyses of Surface and Variable Depth Samples from Tank 43H," WSRC-TR-2001-00073, Rev. 0, March 29,2001
   

Procedure (See attached Tables)

1. Prepare the indicated amount of 50 wt. % NaOH solution and store in PE bottle.
2. In a 2L PE bottle, place approximately 0.8L of DDI H2O and the indicated amount of Ultrex nitric acid.
3. Stir until well mixed and cool to room temperature.
4. Add the indicated amount of Al(NO₃)₃.9H₂O, Add the indicated amount of enriched uranium stock solution.
5. Slowly add the indicated amount of the 50 wt. % NaOH solution with vigorous mixing until a white precipitate persists. Add the remainder of the NaOH solution rapidly with vigorous mixing.
6. Add the remainder of the salts, dilute to mark with DDI water and mix for a minimum of 24 hours.
7. Filter through a Whatman filter paper to remove undissolved solids. Limit exposure to air during filtration.
8. Store solution in tightly stoppered PE bottle labeled with Name, date prepared, concentration and safety hazard warning after flushing vapor volume with nitrogen or argon.

Preparation of 50 wt. % NaOH Solution

(Prepare excess 10% of amount needed to allow for filtration losses).

1. Slowly add indicated amount of NaOH to that indicated for DDI water. Record weight of NaOH and water used.
2. Mix until all solids have dissolved.
3. Filter solution through Whatman filter paper and store filtrate in clean PE bottle labeled as follows.
4. Flush vapor volume of storage bottles with argon before closing and after each opening of bottle.

Appendix B2: Tank 43H Real Waste Tests

Appendix B1
Simulant Analytical Results
Addition Rate Tests

Utility Test

Mixing Tests

Appendix B2
Real Waste Analytical Results

*Lims# refers to analytical tracking number at the Analytical Development Section of SRTC.
**Neat and unfiltered sample values not included in averages.