Kamat, Hrishikesh. Rare-earth oxides in aluminoborosilicate glasses and their impact on molybdenum oxide solubility in nuclear waste glasses. Retrieved from https://doi.org/doi:10.7282/t3-mrz4-1s59
DescriptionThe United States department of energy (DOE) has proposed an alkali alkaline-earth aluminoborosilicate-based glass-ceramic to immobilize the projected high-level waste (HLW) generated from reprocessing of its spent civilian nuclear fuel. The glass-ceramic is expected to improve molybdenum oxide (MoO₃) retention, which otherwise suffers from poor solubility in borosilicate-based glasses typically used for HLW immobilization in several countries. Irrespective of the waste form chosen, a fundamental understanding of the compositional and structural drivers governing MoO₃ solubility in borosilicate/aluminoborosilicate glasses is desired to design advanced waste forms with higher loading capacities. The literature in this regard reports several studies, yet several open questions need to be addressed. This work addresses two open questions regarding MoO₃ solubility: (i) What are the solubility and retention limits of MoO₃ in aluminoborosilicate glasses as a function of glass chemistry? (ii) Why does MoO₃ exhibit significantly higher solubility with the incorporation of rare-earth oxides (RE₂O₃) in aluminoborosilicate glasses?
A systematic study conducted on a series of model HLW glasses reveals that MoO₃ solubility improves by 2x from 1.5 mol% to 3 mol% when RE₂O₃ (RE = Nd) is added to a Na₂O-CaO-Al₂O₃-B₂O₃-SiO₂ glass. The results, when analyzed in the context of past literature, reveal that (i) RE₂O₃ phase separates a homogenous aluminoborosilicate-based glass into borate-rich, and aluminosilicate-rich regions and preferentially enters the borate-rich region; (ii) the excess RE³⁺ clusters in the aluminosilicate-rich region and (iii) molybdenum enters the RE-borate-rich region as the molybdate oxyanion (MoO₄²⁻) forming a stable Mo-RE-B-O glass phase which suppresses crystallization of alkali/alkaline-earth molybdates and improves MoO₃ solubility. The above hypothesis is further explored by investigating the partitioning and clustering behavior of RE³⁺ (RE = Nd/La) in a peralkaline aluminoborosilicate glass doped with varying concentrations of RE₂O₃ (0.001 to 5 mol%). In these glasses, free induction decay (FID)-detected electron paramagnetic resonance (EPR) reveals that RE³⁺ co-exists as EPR-detectable – isolated RE³⁺ centers & dipole-coupled RE clusters, and EPR-undetectable exchange-coupled RE clusters, with higher RE₂O₃ concentrations further promoting RE³⁺ clustering. The environment of the EPR-detectable RE as investigated by electron spin echo envelope modulation (ESEEM) spectroscopy reveals an alkali/silica-rich environment for the isolated RE³⁺ centers and an alkali/boron/silica-rich environment for the dipole-coupled RE clusters. The EPR-undetectable RE clusters are predicted to exist in alkali/boron-rich nano-scale regions, depleting the leftover glass of the same elements. Based on these findings, a study is eventually performed to investigate further the role of RE³⁺ in improving the solubility of MoO₃ in alkali alkaline-earth aluminoborosilicate-based model high-level waste glasses. It is thus hypothesized that MoO₃ preferentially enters the alkali/boron-rich environment of the exchanged-coupled RE clusters where the molybdate oxyanion (MoO₄²⁻) achieves its charge neutrality primarily from RE³⁺ than alkali ions, thereby suppressing the crystallization of the alkali molybdate phase and improving MoO₃ solubility.
A subsidiary study investigating the impact of ruthenium oxide (RuO₂) – one of the components of HLW, on the crystallization behavior and electrical conductivity of MoO₃-containing nuclear waste glasses is also presented. RuO₂ in these glasses is found to exhibit very low solubility (460 parts per million by weight), and above the solubility limit is observed to precipitate into polyhedral/needle-shaped RuO₂ crystals, which impart metal-like conductivity in the investigated glasses.