The role of pe, pH, and carbonate on the solubility of UO 2 and uraninite under nominally reducing conditions
Experimental data obtained from uranium dioxide solubility studies as a function of pH and under nominally reducing conditions in a 0.008 mol dm -3 perchlorate medium and in a 1 mol dm -3 chloride solution are presented. The solubility of extensively characterized uraninite samples from Cigar Lake (Canada), Jachymov (Czech Republic), and Oklo (Gabon) was determined in a solution matching the composition of a groundwater associated with granitic terrain. The redox potential of the test solution was monitored throughout the experimental period. The results obtained were modeled using aqueous formation constants compiled by the NEA, using stability constants corrected to appropriate ionic strengths. The solubility curves have been adjusted by calculating the value of K s4 (UO 2(s) + 2H 2O ⇔ U(OH) 4(aq)) that gave the best fit with the experimental data. For a low temperature synthetic UO 2, a value of logK s4 of -7.3 was determined, while for uraninites the best fit was obtained with a value of logK s4 of -8.5. A wide range of published UO 2 solubilities can be reproduced by the available database, where experimental conditions were adequately defined in the original experiments. A lower value of the solubility product of the uranium dioxide phase defined as fuel in the SKB uranium database provides reasonable solubilities for a wide span of experimental results at near to neutral pH. Based on the modeling and using the β 1,4 for the U(IV)-OH complexation given by Grenthe et al. (1992a), a logK s0 (UO 2(s) + 4H + ⇔ U 4+ + 2H 2O) value of -2.3 ± 0.2 is proposed. Differences in solubility between natural and synthetic samples are attributed to the presence of carbonate in the experiments performed with uraninites, while differences in solubility observed among the natural samples can be correlated to radiation effects at atomic scale.