Understanding uranium fate in wetland soils: a speciation and labile behavior study in the former extraction mine of Rophin (France)

Uranium (U) mining and milling activities, as well as mineral processing plants, raise environmental concerns due to the possible release of radioactive and other potentially toxic elements. To understand their fate in the environment and evaluate their potential impact, the main scientific challenge calls for identifying their solubility, mobility and bioavailability in the environment. Around former U mining and processing plants, wetlands prove to be specific zones with significant amounts of U. This is partly explained by the reduction of the mobile U(VI) into U(IV) due to strongly reducing conditions related to the microbial activity and/or by complexation with the organic matter occurring with high concentrations in wetlands.
At the center of the ancient mining district of Lachaux in France (45.994°N, 3.596°E), the site of Rophin (within the ZATU: Uranium Working Zone = Long Term Socio-Ecological Research Tool of CNRS, Fig.1.a) is characterized by a wetland area with large U concentrations up to 16 g.kg-1 of dry mass of soil [1]. Several cross-analyses indicate that U was transported in particulate forms into the wetland during the exploitation of U(VI) phosphate minerals [1]. The Rophin site therefore provides the opportunity to study the stability of these U minerals over almost 70 years in a non-manipulated wetland since the closure of the mine. In this context, the main challenge is to describe the behavior of U (and decay products of interest) in the wetland using a predictive model that combines transport and chemical speciation. The overall adopted scientific approach is to propose a mechanistic description of the mobility of these elements, from the molecular scale (speciation) to in natura behavior (lability), by coupling field investigations and laboratory experiments.
A simplified three-layer model describes the soil profile of the Rophin wetland, with variable U concentrations and specific physico-chemical soil properties (Fig.1.b). Analyses carried out by X-absorption spectroscopy on in natura soil samples mainly indicate the presence of an adsorbed form of U(IV) in the highly contaminated layer, while both U(IV) and U(VI) are identified in the organically rich part at the surface. Additional SEM-EDX and μ-XRF mapping measurements also indicate that U minerals transported into the wetland were partially dissolved and re-adsorbed in soils.
The objective was to determine which fraction is finally labile, i.e. the fraction that is adsorbed and whose resupply in solution is rapid. This was assessed by desorption experiments under representative site conditions with two complementary approaches [2,3]. Overall, the available fraction is very low (3 to 10 % for the highly contaminated layer and less than 1% for the surface layer) and is characterized by distribution coefficient (Kd) values of the order of 102 L.kg-1. These results were then compared with field data using DGT/DET techniques coupling (Diffusive Gradient in Thin-films/ Diffusive Equilibrium in Thin-Films) [4]. We found an available fraction in the whitish zone, which is rapidly eliminated with time, whereas a labile U fraction is almost undetectable in the surface layer. Combined with time-dependent deployment DGT/DET coupling, results highlighted the predominance of a kinetic parameter and thus, the expected key role of the organic matter in the U mobile behavior in the surface soil.
This study provides access to the input data (labile quantity and (Kd) values) for the establishment of a reactive transport model at the scale of the Rophin mining-affected wetland, but also allows to determine significant information on U interactions in soils. As such, this multi-scale and interdisciplinary study contributes to the improvement of the global understanding of U migration and fate in wetland soils.