Development of a selective americium separation process using TPAEN as a water-soluble stripping agent

International audienceRecycling americium from spent fuels is an important option considered for the future nuclear fuel cycle as americium is the main contributor to the long-term radiotoxicity and heat power of the ultimate waste. The separation of americium alone from a PUREX raffinate can be achieved by co-extracting lanthanide (Ln) and actinide(III) cations in an organic phase containing TODGA diglycolamide extractant, and then strip Am(III) with selectivity towards curium and lanthanides (by analogy with the i-SANEX process for example). The water soluble ligand TPAEN (N,N,N',N'-tetrakis[(6-carboxypyridin-2-yl)methyl]-ethylenediamine) was tested to selectively strip Am from a loaded organic phase. Used in combination with TODGA in the organic phase, TPAEN shows a quite high Cm/Am selectivity (SFCm/Am = 4 at 0.1M HNO3), which allows to selectively strip Am while Cm and Ln remain extracted in the organic phase. The molecule can be used at pH 1 without addition of nitrate salts.Additional batch data were acquired to evaluate the best conditions to develop a liquid-liquid separation flowsheet with this promising TODGA/TPAEN separation system. It was observed that with macro concentrations of lanthanides, an increase of the TPAEN concentration leads to higher Cm/Am selectivity (SFCm/Am up to 5 were measured). It was also demonstrated that TPAEN has a strong complexing capacity allowing the back extraction of Am in macro concentration (1 2 mM) even with low ligand concentrations (around 1.5 equivalent of ligand is enough). TPAEN has a higher affinity for light lanthanides and the selectivity between La and Am might become low depending on the experimental conditions. The influence of parameters such as temperature, pH, ligand and cations concentrations was studied. Kinetics experiments in batch contactors (centrifuges or mixer settlers) were also performed to characterize times necessary to reach equilibrium. This set of experimental work data allowed the elaboration of a thermodynamical model which was implemented in the PAREX simulation code in order propose a flowsheet. The feasibility of this process will be evaluated during spiked tests in centrifugal contactors starting from a surrogate PUREX raffinate spiked with traces amounts of Am and Cm. A final hot test will finally be performed at ITU from a genuine PUREX raffinate.This work is the result of collaborations in the framework of the SACSESS European Project