Optimising the dissolution of U/ Al in alkaline solutions and subsequent dissolution of the generated uranium residue in nitric acid

Abstract

SAFARI-I reactor is a "tank-in-pool" high flux reactor that is situated in Pelindaba, South Africa. The SAF ARI-1 is a material testing reactor that uses LEU target plates, containing 19% uranium enrichment. When producing the target plate (TP), aluminium alloy of nuclear grade (grade 6061 Al sheets) is used. In order to produce the target assembly, a picture-frame method of target assembly is used to produce the target plate by using two aluminium sheets that sandwich another sheet, which has uranium aluminide in it. The uranium aluminide and aluminium powders are amalgamated and compressed to reach the required 2.17 g/cm3 of the LEU fuel density (Ali et al., 2013). During the process of U/ Al target plate production; a lot of scrap material is generated containing uranium that may be recovered, purified and recycled into the target plate manufacturing process. Necsa has followed a process of dissolving the target plate scrap using potassium hydroxide; however, the process has not been optimised. The aim of this study was to determine the optimum dissolution medium for the U/Al scrap material in either NaOH or KOH. Another objective was to evaluate the U recovery process from the residue remaining after alkaline dissolution of the scrap material, and determine if traces of aluminium remaining after dissolving uranium residue in nitric acid had any impact on the recovery of uranium using solvent extraction, and also to assess ion exchange as an alternative purification method. Alkali media concentrations were optimised for both NaOH and KOH, and the optimum concentration, based on the dissolution rate and re-precipitation time, was 6.0 M with a 3.0 M excess alkali concentration required for stoichiometric dissolution of the aluminium. Parameters including stirring rate, temperature and the addition of NaNO3 were also optimised. A combination of all parameters revealed that the optimum dissolution parameters for pure aluminium were an alkaline medium ofNaOH at 6.0 M with 3.0 M excess concentration and 3.5 M NaNO3 and stirring the solution at 1000 rpm. This optimisation was based on the effect each parameter had on both the dissolution rate and re-precipitation time. The optimised parameters were implemented on unirradiated UChem aluminium alloy samples. The optimised dissolution method was an improvement from the previously used method at Necsa. The optimised parameters were further used to dissolve target plates containing depleted uranium (DU). Dissolution kinetics of the DU target plate showed that the dissolution rate of the TP in only sodium hydroxide was higher (0.5328 g/min) when compared to the reaction with the inclusion of sodium nitrate at 3.0 M concentration (0 .1488 g/min). The uranmm residue, from previous dissolutions on site at Necsa was used for optimising dissolution of uranium with varied HNO3 concentrations and temperatures. The optimum dissolution parameters for uranium when favouring a higher dissolution rate were 6.0 M HNO3 at 50°C. Otherwise, when uranium dissolution efficiency is the priority, a lower concentration of 1.0 M HNO3 and lowered temperature of 25 °C are the parameters to follow. Therefore, for this study, a 3.0 M HNO3 concentration at 25°C were chosen as parameters for dissolution of the uranium to be further processed via solvent extraction to recover uranium. This choice was made because tributyl phosphate (TBP) has a lower solubility at that concentration of acid. Solvent extraction was performed using 30% TBP in kerosene with a ratio of aqueous (uranium residue dissolved in nitric acid) to organic being 1: 1. Different amounts of aluminium were added to the uranium solutions to determine the effect the Al concentration has on uranium extraction. Adding Al to uranium solutions had a small effect on extraction; at 1.25% Al (m/m U), the highest extraction of 98.8% was observed and adding any more than that had a negative effect on extraction. Uranium stripping was also investigated by optimising the (NH4)2CO3 concentration between 0.5 M and a higher concentration of 1.0 M. Results showed that at a higher (NH4)2CO3 concentration, uranium stripping was increased from an average of 23 .89% to 83.90%. An alternative uranium recovery method using ion exchange resin was evaluated. Three resins, Amberlite IRC747, Amberlite IRC748 and BioRad AG50W-X4, were tested to determine the one that would recover more uranium. The evaluation was done by comparing the distribution coefficient of uranium on each resin. Uranium had the highest Ko of 4461 .9 ml/dry g with the use of Amberlite IRC747. Amberlite IRC747 was further tested in a column experiment for uranium recovery with a uranium feed concentration of 480 ppm. With the flow rate set to 1.65 ml/min, uranium breakthrough was observed at 60.69 bed volumes, equivalent to a breakthrough capacity of 29 .11 g U/L resin. Comparing both recovery methods, solvent extraction (SX) recovered 86.27% of uranium in the solution, while an ion exchange resin recovered 82.08% uranium. In terms of purification of the uranium, solvent extraction also performed better, since aluminium was removed to below its detection limit, while in the column run no separation of aluminium from uranium was observed.