No abstract

Sellafield Ltd operates a Waste Vitrification Plant (WVP) to immobilise the arisings from the reprocessing of spent nuclear fuel. Washout of solids from the base of waste storage tanks in preparation for decommissioning is likely to produce feeds enriched in molybdenum to the WVP. Vitrification of such feeds in the borosilicate glass formulation currently used by the WVP for vitrification of reprocessing waste has been investigated to determine the maximum achievable loading of MoO3.The vitrification of molybdenum in the absence and presence of reprocessing waste was studied. A number of glasses were manufactured in the laboratory containing various waste loadings. The resultant glasses were examined both visually and under the scanning electron microscope for the presence of any phase separation. Additional aluminium was added to the glasses manufactured in the absence of reprocessing waste to improve the durability of the glass. In borosilicate glass containing 3.5 wt% Al2O3 the onset of a molybdenum phase separation was observed in glasses containing 2.6 wt% MoO3. In the presence of Magnox reprocessing waste, phase separation was observed when the product contained >3.8 wt% MoO3. Soxhlet durability testing of a selection of the glasses manufactured was carried out. The Soxhlet durability of glasses in the absence of phase separation was good.

This paper describes the results from static leach tests using the ASTM International standard Materials Characterisation Centre (MCC-1) and Product Consistency Test (PCT) protocols for inactive High Level Waste (HLW) glasses fabricated at full scale on the Sellafield Vitrification Test Rig. The samples comprised monoliths and powders of a 75:25 Oxide:Magnox Blend glass with 31 wt% waste incorporation and a Magnox-only glass with 35 wt% waste incorporation. The tests were carried out in de-ionized water at 90 °C for durations up to 42 days and normalized mass losses calculated.The results of MCC-1 and PCT tests on both 31 wt% Blend and 35 wt% Magnox glasses, showing measurable differences to the corresponding standard 25 wt% waste incorporation glasses, are presented. A series of Scanning Electron Microscopy (SEM) investigations were also undertaken. The variation in composition and thickness of the alteration layer with sample type and duration is reported.

In the U.K., High Level Waste from reprocessing operations is vitrified at the Sellafield Waste Vitrification Plant (WVP). A small number of the nuclides present in the waste have the potential to volatilize during vitrification. In order to prevent release of any radionuclides to the environment it is important to understand the mechanisms by which volatilization may occur and to have suitable controls in place. One element of particular concern is ruthenium, formed during the fission of nuclear fuel, which has the potential to form gaseous species such as RuO4 during the vitrification process and whose behavior must therefore be understood in order to underpin the safe operation of WVP.

Modeling chemical durability of high level waste glass for nuclear waste processing using bootstrap aggregated neural networks is studied in this paper. In order to overcome the difficulty in developing detailed mechanistic models, data driven neural network models are developed from experimental data. A key issue in building neural network models is that model generalization capability cannot be guaranteed due to the potential over-fitting problem and the limitation in the training data. In order to enhance model generalization, bootstrap aggregated neural networks are used in this study. Multiple neural network models are developed from bootstrap resampling replications of the original training data and are combined to give the final prediction. Application results show that accurate and reliable models can be developed using bootstrap aggregated neural networks.