Milieux Fondus à haute température : nouvelle approche expérimentale et théorique des oxo et fluoroacidités – MILIFOX

The control of acidity is a well-known prerequisite of any chemistry experiment performed in an aqueous solvent. In such liquids, acidity is given by a unique quantity, the pH, which is defined from the activity of the proton. It is readily and precisely evaluated through simple devices or electrochemical methods. In the case of molten salts, the bases that define the acid-base systems are simple anions, for examples O2-- in molten oxides or F- in fluorides. Every species capable of combining with one base is an acid, according to the Lewis definition. Acidity scales are then built up by classifying the strengths with which different acids react with the different bases. As these strengths depend on the nature of both the acid and the base, it is necessary to distinguish between different kinds of acidity. Such scales have already been set up for a variety of molten salts, ranging from molten hydroxides to molten chlorides but two families of molten salts have up to now been largely avoided in such classifications: molten oxides and fluorides. This is due to their strong reactivity and high melting temperature, which makes them difficult to handle properly in the laboratory scale. Nevertheless, they are used in important industrial applications: oxides are importantly used in the glass and ceramics industry, while fluorides have no concurrency for the electrowinning of aluminium. The objective of this project is to build quantitative scales of fluoro- and oxo-acidities within a combined experimental and theoretical approach. The acidity of a molten salt will act directly on the formation of complexes including the various cations of the melt. The determination of the speciation of some given melts with varying overall composition therefore provides an indirect measure of the acidity. This opens a route for building up the acidity scale through the use of some specific in situ spectroscopic approaches. To determine the speciation in the liquid state, Nuclear Magnetic Resonance (NMR) appears as the ideal candidate, as it allows estimating the relative concentration of the various complexes formed in the melt Thanks to new developments NMR will also provide dynamical information through the measurements of self diffusion coefficients. In parallel, an exact characterization of the species formed has to be performed. This can be done by an analysis of the vapour in equilibrium with the melt, for which the preferred technique is the Mass Spectrometry (MS).Because of the atomic scale of the study, theoretical calculations appear as an ideal complement to these experimental techniques. As the systems dealt with are liquids at high temperature, the Molecular Dynamics (MD) simulations provide an adequate level of description. This project proposes a unique combination of NMR, MS and MD in order to set up the acidity scales. As a few amount of dissolved oxide can often be found in molten fluorides, the study will be extended to the so-called oxifluoride systems, which will contain both types of anions. Almost any application based on the use of molten salts (aluminium or ceramic production, nuclear or solar reactor coolants) involves an interface between this salt and a metallic material. In order to avoid the corrosion of the metal surface, the electrochemical properties of the media have to be perfectly controlled. The last point of interest concerns the medium and long-range dynamics of the species in the melts. These properties will affect a lot the physical chemistry of the overall liquid, for example by inducing important changes in the viscosity. The motions of atoms and groups of atoms that are usually discussed in the context of transport properties such as diffusivity and viscous flow in melts are also fundamentally linked to equilibrium properties as heat capacity, entropy and thermal expansion. One of the main efforts in this proposal will be to develop diffusion coefficient measurements at high temperature in these molten media. Here again, a multi-technique approach will also be favoured, since NMR, MD and electrochemistry are able to provide independent estimates of these quantities. Combining these techniques will allow investigating these difficult media from the microscopic properties ' structure and dynamics ' to the macroscopic ones and from the individual to the collective ones. Moreover, they can provide common values that allow testing the relevance of the measured parameters. The unique combination of NMR, MS, Electrochemistry and MD proposed is therefore a powerful tool to set up the acidity scales. It will be associated with an important development of experimental and theoretical tools: high gradient diffusion NMR probe and very high temperature NMR probe, specific electrodes, Mass Spectrometry devices and numerical models.