Assessing the risks of the migration of radiotoxic or chemical markers in the environment relies on our ability to predict the behavior of these pollutants in complex environments where physico-chemical conditions can vary in time and space. Knowledge of the chemical reactions in solution and at solid/solution interfaces must implicitly be linked to the transport properties of the medium. A detailed understanding of the behavior of radioelements in natural environments is therefore essential for the development of predictive reactive transport codes. In real cases of radiological and/or chemical contamination of groundwater and rivers, the source term is generally complex. Interactions between radioelements (“cocktail” effect) can alter their retention properties on solid phases in the surrounding medium. Similarly, the physico-chemical conditions in the environment will influence the speciation of elements in solution, and in particular, their retention properties on reactive phases.
To improve the knowledge of radioelement behavior in soil and groundwater, particularly in a multi-contaminant source (e.g., U, I, Cs, Sr, Ru, Tc), it is essential to understand the behavior of individual radioelements. Ruthenium (Ru), for example, has been identified in the literature as either strongly or weakly mobile, depending on the physico-chemical context. Its behavior in the environment in particular remains poorly understood. In these contexts, Ru occurs primarily in oxidation states +2 to +4, which vary with three main factors: pH, redox potential, and the presence of complexing ligands in solution. To predict the speciation of Ru in natural waters, it is necessary to have complexation constants for the dominant ions in the environment, including ammonium (NH4+), carbonate (HCO3-/CO32-), chloride (Cl-), sulfate (SO42-), nitrate (NO3-), hydroxide (OH-), and phosphate (PO43-). However, equilibrium constants for ruthenium (II-IV) complexes with ligands present under natural conditions are highly variable and limited in the literature. Depending on its speciation in solution, ruthenium can also sorb to or co-precipitate on reactive mineral phases such as clays and carbonates. This chemical reactivity, which depends on the physico-chemical context, is essential for predicting the migration of Ru and other radioelements in the environment.
This thesis aims to fill the gaps in thermodynamic data (complexation in solution, sorption, etc.) for the geochemical modeling of radioelements of interest (in particular ruthenium and technetium) in a natural physicochemical context. It also aims to assess the competitive effects on sorption, both with respect to anions and cations in solution and to mineral phases in the solid medium. This work will include an experimental and geochemical modelling approach.

Radium-226 is one of the main radionuclides remaining in uranium mining residues. However, its direct descendant, radon-222, is a noble gas with a half-life of 3.8 days, which is potentially dangerous to humans if inhaled. In order to minimize the release of this element into the air, mining residues are placed under barriers that limit the diffusive transport of Rn-222 to the surface. The design of these barriers (thickness, materials, water saturation, etc.) should be based on experimental data that quantitatively describe the mobility of radon within them. However, due to the many experimental difficulties associated with studying this radioactive gas, this type of data is rare and often specific to a particular study site, making it difficult to generalize (Fournier et al., 2005; Furhman et al., 2023). However, new investigation techniques have recently emerged that should enable us to deepen our understanding of the diffusive behavior of radon. For example, new devices have been developed to study the diffusion of radionuclides through materials that are partially saturated with water (Savoye et al., 2018; 2024). In addition, spectroscopic autoradiography has recently made it possible to quantify and map alpha emitters present in materials, particularly those in the 238U decay chain and therefore 222Rn (Lefeuvre et al., 2024).
The objective of this doctoral project is therefore to combine these two new approaches to investigate how the diffusion of radon through materials considered as barriers (laterites, bentonite, etc.) can be impacted by the key parameters used to design barriers, namely their degree of water saturation, their level of aging, and their intrinsic heterogeneity.