Dissolution rates of unirradiated UO2, UO2 doped with 233U, and spent fuel under normal atmospheric conditions and under reducing conditions using an isotope dilution method

The experimental results given in this report allow us to draw the following conclusions. 1) Tests using unirradiated fuel pellet materials from two different manufacturers gave very different dissolution rates under air atmosphere testing. Tests for fragments of pellets from different pellets made by the same manufacturer gave good agreement. This indicates that details of the manufacturing process have a large effect on the behavior of unirradiated UO2 in dissolution experiments. Care must be taken in interpreting differences in results obtained in different laboratories because the results may be affected by manufacturing effects. 2) Long-term tests under air atmosphere have begun to show the effects of precipitation. Further testing will be needed before the samples reach steady state. 3) Testing of unirradiated UO2 in systems containing an iron strip to produce reducing conditions gave [U] less than detection limits (<0.02 ppb) after a few days to a few weeks of testing. Uranium recovered from the rinsing of reaction vessels and from acid stripping of vessels was shown to be from dissolution of grains of solid dislodged when the samples were handled after the tests were terminated. 4) Batch tests conducted under reducing conditions showed evidence of colloidal material in the early solution samples. 5) In the batch tests, measurements taken at day 3 and day 5 show that precipitation occurs from day 3 to day 5 without any further dissolution of the solid. 6) At termination of the batch tests, all but one sample had [U] in solution less than detection limits (< 0.02 ppb). Materials recovered in test termination samples showed evidence for recovery of small amounts - amounts corresponding to that expected from a few grains of 5 to 10 μm size - in the acidified solution samples. These are interpreted to have been dislodged during sample handling operations. 7) Batch test data show that increasing test duration beyond 2 weeks does not provide additional meaningful data. 8) A test procedure that used several short exposures of the sample to solution - the puff test procedure - gave results that showed very little recovery of the spike solution at the end of the tests. Only 10% of the 235U added as spike was recovered, indicating that 90% of the spike had precipitated onto the solid sample or the iron strip. 9) Tests of UO2 pellet materials containing 233U to provide an alpha decay activity similar to that expected for spent fuel 3000 and 10,000 years after disposal showed that the pellet materials behaved as expected under air atmosphere conditions, showing that the manufacturing method was successful. 10) Early testing of the 233U-doped materials under reducing conditions showed relatively rapid (30 minute) dissolution of small amounts of U at the start of the puff test procedure. Results of analyses of an acidified fraction of the same solutions after 1 or 2 weeks holding indicate that the solutions were inhomogeneous, indicating the presence of colloidal material or small grains of solid. 11) Samples from the 233U-doped tests initially indicated dissolution of solid during the first week of testing, with some indication of more rapid dissolution of the material with the higher doping. 12) The second cycle of testing of the 233U-doped materials also showed dissolution occurring during the dilution stages of the puff test. The subsequent week of testing also showed small amounts of further dissolution, with hints that the doped samples were dissolving faster than the undoped samples. 13) At the end of 2 weeks of cycle 2 the remaining solution and solid was transferred to a new reaction vessel, the solution was made up to original volume, and a new dose of spike was added. The results of analyses of [U] and isotopic composition show that the measured U is that expected from dilution of the original solution plus adding the spike. 14) Samples taken during 2 weeks of testing of respiked solution showed precipitation of U without further dissolution of the solids. 15) A new dose of spike was added to the remainder of the respiked cycle 2 solutions. [U] and isotopic composition was measured after spiking. Calculations showed that the [238U] before spiking had been about 0.006 ppb. 16) In the test termination after 7 days, all solution samples were < 0.02 ppb, indicating precipitation of the spike. About 50% of the precipitated spike was recovered in acid stripping of the vessels. 17) Removal of U from the iron strips with sodium bicarbonate solution only recovered 5% of the missing spike. Thus, considerable amounts of the precipitated spike U seem to have gone onto the surface of the test samples. 18) In the final tests with 233U-doped materials there was no evidence of enhanced dissolution due to alpha radiolysis. 19) Testing with spent fuel under 10 bar H2 atmosphere showed decreasing [U] together with increasing [Cs] indicating that the H2 was able to act as a reducing agent in the system. 20) Addition of a 235U enriched spike caused larger than desired increases in [U]. Despite the rapid precipitation that subsequently occurred, we were able to calculate a dissolution rate for U from the spent fuel that was occurring even in the presence of precipitation. 21) Further testing is needed using more sensitive analytical techniques to establish the dissolution rate of the 233U-doped materials. 22) Further testing of spent fuel is needed to verify the dissolution rate under H2 atmosphere conditions and to obtain a dissolution rate in the presence of actively corroding iron