The recycling of metals by hydrometallurgy is necessary to ensure both the development of new energy technologies and the sustainability of the nuclear power cycle. The industrial processes used here, such as liquid-liquid extraction, involve the flow of two fluids under agitation whose interfaces form and deform. The coalescence of drops thus plays a very important role. The modelling of these complex two-phase systems involved in the extraction process must therefore take this phenomenon into account in the best possible way. In this thesis, we propose to describe for the first time the coalescence of drops in a realistic way by a multi-scale approach to take into account the physicochemical aspects of the phenomena. Firstly, molecular dynamics simulations will allow us to propose a stochastic model where the probability of coalescence will be expressed as a function of geometric parameters (distance and size of drops). Then the role of physico-chemical quantities in the phenomenon will be systematically described (role of surface tension, presence of surfactants, viscosities, etc.). The ultimate goal is to arrive at computational fluid dynamics (CFD) simulations in which coalescence, and in particular its random nature, will have been developed and validated by molecular dynamics calculations performed at the atomic level, taking into account the physicochemistry of the interface. Such a model would be a major step towards predicting the kinetics of liquid two-phase media, not only in the context of optimising recycling processes, but also for many other applications.
Candidate profile: Master of Physics - Master of Chemistry - Engineering School - ENS. After the thesis, the candidate can either continue in the academic field, with a high level of competence in modelling, or move towards industry by developing his/her dual competence in modelling and recycling.