Impact of non‐framework cation mixing on the structure and crystallization behavior of model high‐level waste glasses

Abstract : Spinel crystallization is known to be detrimental to the operation of Joule heated ceramic melters during the vitrification of iron-rich high-level nuclear wastes (HLW) into borosilicate glasses. The literature on this subject focuses on tackling the problem by developing empirical constraints to design compositions, which limit the fraction of spinels formed in the melter or by developing empirical models to predict the settling behavior of spinels in the melter as a function of the glass composition. While these empirical models can predict the behavior of most of the compositions, they are not failsafe as there are always some compositions, whose behavior is beyond the predictive ability of these models. This can lead to undesirable situations during the vitrification of the nuclear waste, and therefore an in-depth investigation of the chemo-structural descriptors controlling the crystallization behavior in these glasses is warranted. Accordingly, the present study aims to understand the impact of non-framework cation mixing (i.e., Li+/Na+ and Ca2+/Na+) on the structure (through Raman spectroscopy and Mössbauer spectroscopy) and crystallization behavior (through XRD, SEM-EDS, and vibrating sample magnetometry) of iron-rich model HLW glasses in the system: (mol.%) x MyO–(25−x) Na2O–9.12 B2O3–6.4 Al2O3–51.25 SiO2–7.22 Fe2O3–0.38 MnO–0.08 Cr2O3–0.55NiO (MyO = Li2O or CaO).