Amorphous Intermediate States during the Precipitation of Cerium Oxalate: Towards a New Model

The precipitation is a common step in industry which is relatively ill-understood. In the last decades, many efforts were made to make numerical simulations of this step to avoid using the trial and error method, especially for the precipitation of active compound like plutonium oxalate. Many parameters are required to handle these numerical simulations, including the nucleation rate (the number of particles which appears in the reaction media by unit of time and of volume).

The nucleation rate is usually rationalized according to the Classical Nucleation Theory (CNT). This theory predicts the nucleation rate, assuming that thermal fluctuations of the reaction media lead to the formation of some “clusters” of monomers with the same composition and (?) symmetry as the crystal. These clusters tend to immediately dissolve into the reaction media unless they reach a particular size, called the critical size, from which they can grow fast to a macroscopic size. The so-called nucleation rate is the rate at which clusters reach this critical size. However, the CNT is more and more questioned, as it has been evidenced for many systems, including cerium oxalate, that nucleation occurs through amorphous states which are currently ignored by the CNT and which lead to an significant underestimation of the nucleation rate, up to 400 orders of magnitude.

In this work, the nucleation of cerium oxalate, a non-radioactive surrogate for plutonium oxalate, has been studied using cryo-microscopy and small and wide x-ray scattering (SAXS and WAXS). Here we show that cerium oxalate has two possible non-classical nucleation mechanisms: at low concentrations, nucleation in presence of amorphous nanoparticles, and at higher concentration, nucleation involving both a transient dense liquid phase and amorphous nanoparticles.