Approche Multi-échelles du Partage de soluté entre phases LIquides – AMPLI

Liquid/liquid extraction is a separation process with multiple applications: in chemistry laboratories; in chemical industry, for the production of fine organic chemicals; for cleaning up techniques; in hydrometallurgy (processing of ores or reprocessing of nuclear wastes). The complexity of these two-phase systems and their high ionic content makes their description difficult, requiring fundamental studies to allow for predictive simulations. In this work, for the sake of practical realism, we propose to focus our attention on the description of the equilibrium thermodynamic and structure aspects. The features concerning the kinetics and supramolecular organisation, and the precise study of the interface, will be reserved for subsequent work. The real systems envisaged will be those of interest for the reprocessing of nuclear wastes, for the following specific reasons: • This part of hydrometallurgy is, in France, of first importance, from the energetic and economic viewpoint. It is one of the last research topics in hydrometallurgy in France. The results will allow one to build up the simulation of separation operations for all actinides. • For reasons particular to this institution, a fundamental study of these systems could not be developed in the CEA (Commissariat à l'Energie Atomique), despite earlier intentions of various researchers (e.g., Claude Poitrenaud et Claude Musikas). This project requires the contribution of various researchers, coming from the CEA and the academy (university, CNRS). Let us note that this systematic study has not been undertaken before in a foreign country (like Russia or USA). Only more or less empirical descriptions have been devised, e.g. by the famous A.M. Rozen in the USSR in the 70's and 80's. It is the ambition of taking into account all physical-chemical features of a solvent extraction system in order to describe and, one day, to predict, its equilibrium properties which constitutes the main scientific originality of the present project. We plan the separate examination of the two phases: 1. For the aqueous phase, it is planned to tackle the problem of departures from ideality in concentrated ionic solutions, which is the main issue. The tools needed for the representation of the thermodynamic properties of the aqueous phase (composed of water, metallic solute and generally an acid) are basically known. The modelling will be achieved by following two approaches: the theory of 'simple solutions' and the BIMSA (BInding Mean Spherical Approximation) model. The advantage of the first approach is its outstanding efficiency/performance ratio. However, this empirical tool does not permit correlations to be made with microscopic and structural features of solution. The BIMSA model is more involved, with more complex, but analytic, equations. On the other hand, it is expressed in terms of quantities that have some physical meaning, allowing for a much more direct connection with microscopic data (experimental or computed). The originality of this part will lie in the establishment of a link between this BIMSA representation and numerical molecular dynamics simulations. By extending classic statistical mechanical theories (the Barker & Henderson or WCA theories) to the present case, these simulations will be used to quantify the parameter values of the BIMSA model and to deepen our knowledge of the structure of the concentrated solutions. The results will be tested by appropriate experimental validations. 2. For the organic phase, the objective is to obtain its structural description and, on this basis, to achieve its thermodynamic representation. The nature of the species and the metal-ligand bonds will be studied using various techniques: mass spectroscopy with an electrospray ionization mode, time-resolved laser spectroscopy (TRLS), vibrational spectroscopy (FTIR) and cyclic voltammetry. Microcalorimetry will be a complementary technique. The role of water molecules will be a key feature of our studies (TRLS). A theoretical study will also be undertaken to identify the stable species in these solutions and to verify the vibration modes obtained by spectroscopy. Let us underline that it is the detailed experimental examination of the organic phase which constitutes the main originality of this work. In contrast to earlier works, we regard this study as an indispensable prerequisite for a satisfactory modelling of the thermodynamic properties of that phase. On this basis, it will be possible to model the extraction equilibria. This work will pave the way for non-empirical models adapted to other liquid/liquid extraction systems.