WSRC-MS-99-00860
Tc-99 and Cs-137 Volatility from the DWPF Production Melter during
Vitrification
of the First Macrobatch at the Savannah River Site
N. E. Bibler, T. L. Fellinger, S. L. Marra, R. J. O’Drisscoll,
J. W. Ray, and W. T. Boyce
Westinghouse Savannah River Company
Aiken, S.C. 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
Abstract
Technetium-99 and cesium-137 are two radionuclides in high level waste (HLW) that can volatilize from high temperature melters during the immobilization of the HLW into a borosilicate glass. At Savannah River Site (SRS) we have obtained data that indicate that this volatilization is small from the full scale production melter in the Defense Waste Processing Facility (DWPF). These data were obtained during the vitrification of the first HLW macrobatch at SRS. This campaign lasted ~2.5 years and produced ~9 hundred metric tons of glass from ~1.6 million liters of HLW. Losses of Tc-99 and Cs-137 were determined by comparing their measured concentrations in the glass with their respective predicted concentrations based on the composition of the HLW being vitrified. For three glass samples taken during the campaign, the measured and predicted concentrations agreed within 7% or better indicating a small loss of either radionuclide. The DWPF melter operates with a cold cap on the surface of the melt. This cold cap could enhance the radionuclide retention, especially Tc-99.
Introduction
The DWPF at the SRS has completed vitrifying the first macrobatch of radioactive HLW sludge slurry into a stable borosilicate glass. This HLW is composed of U-235 fission products and radioactive actinide isotopes produced in the reactors at SRS during production of special nuclear material for the U.S. Government. However the major constituents of the HLW are nonradioactive elements used in the processes at SRS to purify the special nuclear material. The DWPF process converts the caustic HLW slurries into a borosilicate glass by mixing the pretreated slurry with a glass forming frit and then feeding this mixture to a joule-heated melter at 1150°C. The molten glass is then poured into stainless steel canisters that are 3m tall and 0.6m in diameter. The canister is then sealed by welding a plug in the top of the canister. It is then stored prior to shipment to a Federal Repository. During the vitrification process there are two U-235 fission products, Tc-99 and Cs-137, which have the possibility of volatilizing from the melter. For example, Tc-99 forms an oxide (Tc2O7) that has a boiling point of approximately 311°C.[1] An extensive study by Vida [2] has shown that substantial amounts of Tc-99 can be volatilized during vitrification in laboratory melters and in a large scale melter. Further, when both Tc and Cs are in the waste, it has been shown nominally 60% of the Tc and 10% of the Cs were volatilized in laboratory scale melter tests. [3]
During the processing of Macrobatch One in the DWPF, data were obtained that allow a determination of the losses of Tc-99 and Cs-137 from the DWPF production melter. These determinations were made by comparing the measured concentrations of Tc-99 and Cs-137 in three glass samples taken at different times throughout the campaign with predicted concentrations based on measurements of the Tc-99 and Cs-137 concentrations in the HLW being vitrified. The results of the determinations are presented in this paper. Direct quanitative measurement of the Tc-99 and Cs-137 losses by analyzing the melter offgas samples in the DWPF would be extremely difficult due to the equipment configuration in the DWPF.
The Macrobatch One campaign in the DWPF lasted ~30 months from 3/96 to 9/98. Details of the campaign are summarized in Table I. The DWPF production melter holds ~6.6E03kg (~2.5E03L) of molten glass. The top surface area of the melt is 2.6m2 and is 95% covered by a cold cap while the waste-glass frit slurry (~60 wt% water) is being fed to the melter. The feed rate is nominally 2L/min. Note in Table I that there was approximately a 21 month period between the first two and the third glass sample. These samples were taken from the pour stream of the melter using a special removable sampler that fit into the top of a canister and collected ~40 grams of glass. The glass samples were sent from the DWPF to the Savannah River Technology Center (SRTC) where they were analyzed. Samples of the HLW sludge slurry taken from the 3.7E06L DWPF feed-tank (Tank 51) to the DWPF were also analyzed at SRTC.
Table I. Properties of the Macrobatch One Campaign in the DWPF
|
Volume of Sludge Slurry Vitrified |
~1.6 Million Liters |
|
Amount of Glass Produced |
~0.9 Million Kilograms |
|
Number of Canisters Filled |
495 |
|
First Glass Sample Taken During Pouring of Canister 50 |
Taken on 9/2/96 |
|
Second Glass Sample Taken During Pouring of Canister 61 |
Taken on 9/23/96 |
|
Third Glass Sample Taken During Pouring of Canister 409 |
Taken on 6/30/98 |
Experimental
The sludge slurry and glass samples had to be dissolved and analyzed in order to determine their composition. These dissolutions were performed remotely in the Shielded Cells of SRTC using procedures developed and tested at SRTC. The resulting solutions were diluted and removed from the Shielded Cells for analyses by various techniques. For the HLW sludge slurry, a large sample was thoroughly mixed and portion taken for dissolution. This sample was dried at 115°C and four samples were taken from it for dissolution. Each of these samples was dissolved by two separate dissolution methods. These methods were an aqua regia method at 115°C in sealed Teflon vessels and a sodium peroxide fusion followed by a HNO3 dissolution. For the glass dissolutions, the three glass samples listed in Table I were crushed and sieved prior to dissolution. The portion passing through a 200 mesh sieve was used for dissolution of each glass. Quadruplicate samples of each were dissolved by two separate dissolution methods. The first used a mixture of HF/HNO3/HCl/H3BO3 solutions at 115°C in sealed Teflon vessels. The second was a sodium peroxide fusion. The solutions from the acid dissolutions in sealed Teflon vessels were used for the determination of Tc-99 or Cs-137, since no losses of these radionuclides could occur during these dissolution procedures. A standard glass with a composition similar to the DWPF glass, (Analytical Reference Glass-1)[4] was dissolved and analyzed each time a set of sludge slurry or glass samples were dissolved in order to check if the dissolutions were complete and the analytical procedures were performed correctly. For each set of dissolutions, the results confirmed that this was the case.
The major nonradioactive waste elements (excluding oxygen) in the dissolved HLW sludge slurry and the glass samples were measured by analyzing the diluted solutions from the Shielded Cells by inductively coupled plasma emission spectroscopy (ICP-ES). Cs-137 was determined by gamma counting the Ba-137m gamma ray that is in secular equilibrium with Cs-137. Tc-99 was determined by inductively coupled plasma mass spectroscopy (ICP-MS). It is proven below that the response at mass 99 in these solutions was indeed due to Tc-99 and not any other isotope such as natural Ru-99 that could have been in the waste.
The analytical method ICP-MS gives the concentrations of analytes as a function of mass. For radioactive HLW solutions, it has been shown that specific isotopes can be assigned to these masses by considering the natural isotopes in the waste and the U-235 fission products at that respective mass.[5] At mass 99, there are two possibilities – natural Ru-99 and the U-235 fission product Tc-99. It can be proven that the response at mass 99 is from Tc-99 and not Ru-99 by considering the concentrations measured at masses 101, 102, and 104. These masses can only be due to natural or fission product Ru. All the radioactive U-235 fission products with these masses have half lives too short to still be present in the sludge at a reasonable concentration. There are no other natural isotopes at these masses except Pd-104 at mass 104. However Pd-104 is blocked in the isobaric beta decay chain by stable Ru-104. Table II gives the measured concentrations at masses 101, 102, and 104 in the dried HLW sludge slurry. The natural abundances and fission yields of Ru-101, 102, and 104 are also given. The distributions based on natural abundance and fission yields are compared to the measured distribution in the last three columns of Table II.
Table II. Proving that the Isotope Measured at Mass 99 is not Natural Ru-99
|
Isotope |
Natural Abundance |
U-235 Fission Yield |
Measured Concentration |
Distribution |
Distribution |
Measured |
|
Ru-101 |
12.1 |
5.18 |
1.3E-03 |
0.19 |
0.46 |
0.46 |
|
Ru-102 |
31.6 |
4.29 |
1.1E-03 |
0.51 |
0.38 |
0.39 |
|
Ru-104 |
18.6 |
1.88 |
4.0E-04 |
0.30 |
0.16 |
0.14 |
The measured distribution clearly fits the distribution of fission yields indicating the isotopes measured at these masses are fission product Ru and not natural Ru. The response at mass 99 must then be due to Tc-99 which, with its long half life (2E05 years), blocks the Ru-99 in the beta decay of the isobaric chain at mass 99.
Results and Discussion
Pertinent Concentrations in the Dried Sludge Slurry and in the Three Glass Samples
The radioactive and nonradioactive composition of the HLW sludge slurry that was vitrified as Macrobatch One was determined at SRTC prior to processing this material in the DWPF. An example of this composition for Tank 51 dried sludge slurry is given in Reference 5. The elements Fe, Al, Mn, and Ca, are the major nonradioactive elements (excluding oxygen) in the HLW that are solidified in the glass. They comprise ~88 elemental percent of the dried sludge slurry. The other major nonradioactive elements are Na and Mg (8%), but these are also present in the glass forming frit and thus their concentrations are not specific to the HLW in the final glass. Since Fe, Al, Ca, and Mn are the major elements in the HLW, the concentrations of these were used in the method to determine the losses of Tc-99 and Cs-137 from the DWPF production melter. Table III shows the measured concentrations of Fe, Al, Ca, and Mn in the dried sludge slurry and in the three glass samples taken during processing of Macrobatch One. The measured concentrations of Tc-99 and Cs-137 are also presented. Standard deviations are given for the replicate samples dissolved and analyzed.
Table III. Concentrations (Weight Percent) of Major HLW Elements
and Tc-99 and Cs-137
in the Dried HLW Sludge Slurry and Three DWPF Glass Samples from the Macrobatch
One Campaigna
|
Species |
Sludge Slurry |
Canister 50 |
Canister 61 |
Canister 409 |
|
Fe |
25.8±0.7 |
9.11±0.21 |
8.44±0.37 |
8.78±0.31 |
|
Al |
6.36±0.30 |
2.34±0.16 |
2.13±0.13 |
2.26±0.10 |
|
Ca |
2.39±0.12 |
0.88±0.05 |
0.79±0.04 |
0.93±0.03 |
|
Mn |
2.61±0.12 |
0.88±0.01 |
0.81±0.03 |
0.86±0.03 |
|
Tc-99 |
(1.28±0.06)E-03 |
(4.34±0.20)E-04 |
(4.08±0.28)E-04 |
(4.67±0.26)E-04 |
|
Cs-137 |
(7.17±0.32)E-05 |
(2.62±0.03)E-05 |
(2.49±0.10E-05) |
(2.45±0.09)E-05 |
a For
the dried sludge slurry, the result is the average and standard deviation for
three samples. For the
glasses,eight samples were analyzed except for Tc-99 where only four samples
were analyzed.
The concentrations measured in the three glass samples are in good agreement as expected indicating that the composition of the HLW did not change significantly during processing. Also note the concentrations in the glasses are lower than those in the dried HLW sludge slurry. This lowering is due to the addition of the nonradioactive glass forming frit that dilutes the HLW elements. This dilution is one of the bases for predicting what the concentrations of Tc-99 and Cs-137 should be in the glass if they are not volatilized from the melter.
Method for Predicting Tc-99 and Cs-137 Concentrations in the Glass
If Tc-99 and Cs-137 are not volatilized from the melter, they should be diluted by the same factor that causes the dilution of the major HLW elements. This presumes that Fe, Al, Ca, and Mn are not lost in the vitrification process. This has been shown to be the case by comparing the compositions of vitrified samples from two melter feed tank batches (measured in the DWPF process control analytical laboratory) with the respective compositions of the glass being poured (Canisters 50 and 61) while those batches were in the melter feed tank.[6] The agreement was excellent for the major nonradioactive elements and for Cs-137 (Tc-99 could not be measured in the DWPF process control analytical laboratory.). This waste dilution factor (WDF) is the concentration of an element in the glass divided by its respective concentration in the dried sludge slurry. Table IV shows the WDF calculated in each glass based on each of the major elements in the HLW. The average, standard deviation, and percent relative standard deviation for each glass sample are also given in Table IV.
Table IV. Waste Dilution Factors (WDF) Based on Major Waste
Elements in
Three DWPF Glass Samples in the Macrobatch One Campaign
|
Element |
Canister 50 |
Canister 61 |
Canister 409 |
|
Fe |
2.83 |
3.06 |
2.94 |
|
Al |
2.71 |
2.98 |
2.81 |
|
Ca |
2.71 |
3.02 |
2.58 |
|
Mn |
2.96 |
3.21 |
3.05 |
|
Average |
2.80 |
3.07 |
2.84 |
|
Std. Dev. |
0.12 |
0.10 |
0.20 |
|
Percent RSD |
4.1 |
3.2 |
7.2 |
The final averages for the three glasses are in good agreement with the relative precision of each average being better than 8% based on the four elements. The predicted concentration of Tc-99 and Cs-137 in each of the glasses is then its respective concentration in the dried HLW sludge slurry divided by the appropriate WDF. These results are presented in the next section.
Comparison of Measured and Predicted Tc-99 and Cs-137 Concentrations in the Glass
Table V shows the measured and predicted and concentrations of Tc-99 and Cs-137 in each of the three glass samples taken during processing of Macrobatch One. The percent difference for each was calculated from the formula
% Diff. = ((Pred. Conc. – Meas. Conc.)/ Pred. Conc.) X 100 (1)
It can be seen that in all cases the predicted and measured concentrations are in close agreement suggesting little volatilization of the Tc-99 or Cs-137. In those cases where a negative difference is indicated, the measured concentration was larger than the predicted. This is due to experimental errors in measuring the respective concentrations.
Table V. Measured and Predicted Concentrations (Wt. %) of Tc-99
and Cs-137
in Three DWPF Glass Samples in Macrobatch One Campaign
It should be mentioned that the results in Table V may be prone to a large relative error because we are trying to determine a small difference between two relatively large numbers. For example, for the glass in Canister 409, the minimum concentration predicted for Tc-99 in the glass is 4.01E-04 wt.% while the maximum is 5.08E-04 wt.%. This is a difference of 20-25% and is due to just the measured precision of the average concentration of Tc-99 in the waste and the precision of the average WDF presented in Table IV for Canister 409. On this basis it is difficult to assign an exact number to the possible losses of Tc-99 and Cs-137; however, based on these data, it appears that these losses are small. As stated earlier, a much better estimate could be made if it were possible to get a direct measurement the Tc-99 and Cs-137 being captured by the DWPF offgas system.
As mentioned before, during feeding of the DWPF melter, the melt pool is 95% covered with a cold cap due primarily by heat transfer from the molten glass to the water from the slurry. The water is evaporated and collected by the offgas condensate tank in the DWPF. This cold cap would clearly decrease the volatility of the Tc-99 and Cs-137. The cold cap is burned off during those times that the melter is not being fed; however, these times were short compared to the times that the melter was being fed. Another possibility for decreasing the volatility of Tc-99 is that the Tc-99 may be in the +4 valance state due to formic acid being added to the sludge slurry in the DWPF as a process chemical in a pretreatment step. The +4 state forms a less volatile oxide that the +7 state. [2]
At SRTC we had the opportunity to make these same type of measurements during a demonstration of the DWPF vitrification process using a remote slurry fed research melter in the Shielded Cells.[7] This demonstration was with the HLW from Tank 51 prior to its washing to be sent to the DWPF. The melt pool of this melter contained only 10kg of glass and the feed rate was so slow that a cold cap never formed.[7] As with Macrobatch One we measured the concentrations of Fe, Al, Ca, and Mn, in the waste and in a sample of the final glass. In this case a WDF of 3.07±0.21 was obtained.[5] The measured concentrations of Tc-99 and Cs-137 in the sludge and glass along with their predicted concentrations are presented in Table VI.
Table VI. Measured and Predicted Concentrations (Wt. %) of Tc-99
and Cs-137 in Tank 51 Glass
Prepared in the Small Slurry Fed Research Melter in SRTC
|
Sludge |
Glass |
|||
|
Isotope |
Measured |
Measured |
Predicted |
Difference |
|
Tc-99 |
1.7E-03 |
3.8E-04 |
5.5E-04 |
31% |
|
Cs-137 |
6.7E-05 |
2.2E-05 |
2.3E-05 |
4.3% |
In this campaign, the results indicate that 31% of the Tc-99 was lost while very little Cs-137 was lost. In this case, loss of the Tc-99 may be due to the absence of a cold cap.
Conclusions
Acknowledgement
This paper was prepared in connection with work done under Contract No. DE-AC09-96SR18500 with the U. S. Department of Energy.
References