CRUCIBLE TESTING OF TANK 48H RADIOACTIVEWASTE SAMPLE USING FLUIDIZED BED STEAMREFORMING TECHNOLOGY FOR ORGANICDESTRUCTION

Abstract

The purpose of crucible scale testing with actual radioactive Tank 48H material was to duplicate the test results that had been previously performed on simulant Tank 48H material. The earlier crucible scale testing using simulants was successful in demonstrating that bench scale crucible tests produce results that are indicative of actual Fluidized Bed Steam Reforming (FBSR) pilot scale tests. Thus, comparison of the results using radioactive Tank 48H feed to those reported earlier with simulants would then provide proof that the radioactive tank waste behaves in a similar manner to the simulant. Demonstration of similar behavior for the actual radioactive Tank 48H slurry to the simulant is important as a preliminary or preparation step for the more complex bench-scale steam reformer unit that is planned for radioactive application in the Savannah River National Laboratory (SRNL) Shielded Cells Facility (SCF) later in 2008. The goals of this crucible-scale testing were to show 99% destruction of tetraphenylborate and to demonstrate that the final solid product produced is sodium carbonate. Testing protocol was repeated using the specifications of earlier simulant crucible scale testing, that is sealed high purity alumina crucibles containing a pre-carbonated and evaporated Tank 48H material. Sealing of the crucibles was accomplished by using an inorganic 'nepheline' sealant. The sealed crucibles were heat-treated at 650 C under constant argon flow to inert the system. Final product REDOX measurements were performed to establish the REDuction/OXidation (REDOX) state of known amounts of added iron species in the final product. These REDOX measurements confirm the processing conditions (pyrolysis occurring at low oxygen fugacity) of the sealed crucible environment which is the environment actually achieved in the fluidized bed steam reformer process. Solid product dissolution in water was used to measure soluble cations and anions, and to investigate insoluble fractions of the product solids. Radioanalytical measurements were performed on the Tank 48H feed material and on the dissolved products in order to estimate retention of Cs-137 in the process. All aspects of prior crucible scale testing with simulant Tank 48H slurry were demonstrated to be repeatable with the actual radioactive feed. Tetraphenylborate destruction was shown to be >99% and the final solid product is sodium carbonate crystalline material. Less than 10 wt% of the final solid products are insoluble components comprised of Fe/Ni/Cr/Mn containing sludge components and Ti from monosodium titanate present in Tank 48H. REDOX measurements on the radioactive solid products indicate a reducing atmosphere with extremely low oxygen fugacity--evidence that the sealed crucible tests performed in the presence of a reductant (sugar) under constant argon purge were successful in duplicating the pyrolysis reactions occurring with the Tank 48H feed. Soluble anion measurements confirm that using sugar as reductant at 1X stoichiometry was successful in destroying nitrate/nitrite in the Tank 48H feed. Radioanalytical measurements indicate that {approx}75% of the starting Cs-137 is retained in the solid product. No attempts were made to analyze/measure other potential Cs-137 in the process, i.e., as possible volatile components on the inner surface of the alumina crucible/lid or as offgas escaping the sealed crucible. The collective results from these crucible scale tests on radioactive material are in good agreement with simulant testing. Crucible scale processing has been shown to duplicate the complex reactions of an actual fluidized bed steam reformer. Thus this current testing should provide a high degree of confidence that upcoming bench-scale steam reforming with radioactive Tank 48H slurry will be successful in tetraphenylborate destruction and production of sodium carbonate product.