The determination of thorium and uranium by reversed-phase liquid chromatography

This thesis presents the results of a systematic study of the reversed-phase high performance liquid chromatographic determination of thorium(IV) and uranyl as their o-hydroxymonocarboxylic acid complexes. Various parameters which affect complexation and retention have been investigated. The aims of this study were to elucidate the retention mechanism of these complexes and use this knowledge for the development of methodology for the determination of low concentrations of thorium and uranium, especially in complex sample matrices. Throughout this study, a conventional reversed-phase chromatographic system was used, which employed a C18 reversed-phase column with aqueous methanolic mobile phases containing a complexing agent. Thorium(IV) and uranyl were detected at 658 nm after postcolumn derivatisation with Arsenazo III. The retention behaviour of thorium(IV) and uranyl a-hydroxyisobutyric acid (RIBA) complexes was firstly investigated. When the percentage of methanol in the mobile phase was varied, both thorium(IV) and uranyl complexes exhibited typical reversed-phase behaviour. Decreased retention was observed when the concentration of MBA in the eluent increased due to competition by the neutral, protonated HD3A in the eluent for the adsorption sites on the stationary phase. Elevation of the mobile phase pH caused the retention of both thorium(IV) and uranyl to increase as a result of the decreased concentration of the neutral ligand acid in the mobile phase. A retention mechanism based on hydrophobic adsorption was therefore suggested. The hydrophobic adsorption mechanism was further confirmed when glycolate or mandelate were used as ligands in the mobile phase. Varying the percentage of methanol, the ligand concentration in the mobile phase, or the pH, showed that both glycolate and mandelate complexes gave similar retention behaviour to that of HlBA. However, the degree of retention observed for glycolate complexes was weak due to the short carbon chain in the ligand, whereas the mandelate complexes were retained much longer because of the increased hydrophobicity arising from the aromatic moeity in the ligand. The elution characteristics of thorium(IV) and uranyl complexes were therefore shown to be dependent chiefly on the ligand hydrophobicity. A reversed-phase chromatographic method was developed for the determination of thorium(IV) and uranyl in mineral sands, using 400 mM HIBA as the eluent. A range of sample dissolution and clean-up procedures was evaluated, with the optimal sample preparation procedure involving a tetraborate fusion and nitric acid leach, followed by either cation-exchange pretreatment, or simple dilution in concentrated HIBA. The results obtained using the chromatographic method showed good agreement with X-ray fluorescence results and gave detection limits about 1.0 lig/m1 for the two analytes. Trace levels of thorium(IV) and uranyl were determined using an on-line preconcentration method which employed a short C18 column as the concentrator. Different ligands were added to the sample solution to enhance retention of the analytes on the concentrator, with 42 mM mandelate being optimal. The sample loading speed was found to have no effect on recoveries if maintained below 5.0 ml/min. A linear relationship between the peak area and the loaded sample volume existed up to 50 ml of sample. The interferences of most common mineral acids and cations were investigated. The on-line preconcentration method, which gave detection limits of 5 ppb , was applied to the determination of trace levels of thorium(IV) and uranyl spiked in sea water. An on-line matrix elimination procedure was then developed as a means of eliminating interference effects and was applied to the determination of thorium(IV) and uranyl in a digested phosphate rock solution. Theoretical calculations predicted that it was impossible to separate thorium(IV) and uranyl from the matrix in one step. Phosphate and other anions were firstly removed from the sample on a cation- exchange precolurnn with a dilute nitric acid eluent. Then thorium(IV) and uranyl were separated from lanthanides on a short C18 precolumn using a mandelate eluent. Finally thorium(IV) and uranyl were transferred onto the C18 analytical column for separation and quantification. The matrix elimination procedure was performed automatically by a programmable HPLC pump.