Liquid precursors in mineral crystallization : an experimental test of nucleation theories

Liquid precursors in mineral crystallization : an experimental test of nucleation theories

November 7 2025
Types d’événements
Thèses ou HDR
NIMBE/LIONS
INSTN
Amphithéâtre Jules Horowitz
November 7 2025
from 14:00 to 17:00

Crystallisation in solution is involved in a wide range of processes, both natural and industrial. Yet, modelling this process remains challenging because it often involves non-crystalline intermediates, and particularly liquid droplets enriched with reactants, that are not accounted for in classical nucleation theory. Alternative theoretical frameworks have been developed to describe such transients, but their experimental validation is hindered by the difficulty of characterizing these droplets : they are nanometer-sized and often live less than a minute, which limits access to their properties and formation mechanisms, even though these are crucial for choosing a suitable model. In this context, the aim of this thesis is to experimentally test current theories on a model mineral system, cerium oxalate, for which a liquid-like transient has been identified. To this end, significant technical and methodological developments were carried out to enable early detection (< 1 ms) of this intermediate, its evolution, as well as the quantification of its fluidic and chemical properties. Thanks to these advances, we (i) demonstrated by liquid-phase electron microscopy the liquid nature of this intermediate, determined its viscosity, and revealed a two-step crystallization process from it, (ii) characterized its chemical composition using cryo-X spectrotomography, and (iii) captured its formation mechanism down to the sub-millisecond timescale by combining microfluidics with in situ X-ray scattering, showing a process consistent with a liquid–liquid phase separation (LLPS) via spinodal decomposition.

These results thus reveal that only a diffuse interface theoretical model is suitable for describing the formation and evolution of this liquid intermediate, but that its application will require further theoretical developments to account for the strong viscosity contrast between the newly formed phases. The experimental approach developed in this thesis can be directly extended to other inorganic systems, thereby paving the way for a unified description of non-classical crystallization through liquid intermediates.

Keywords : SAXS – Microfluidics – Nucleation – Spinodal decomposition – Cryo-TXM