Nepheline crystallization and the residual glass composition: Understanding waste glass durability

Abstract

The vast majority of High‐Level Waste (HLW) originating from defense nuclear programs is sequestered and immobilized in borosilicate glass. Borosilicate glass is universally accepted for immobilizing HLW, but its efficiency has limitations based on the compositional makeup of the waste stream. The chemical durability of the glass is the most important factor in determining the longevity and usefulness of the final glass waste form. The primary detriment to this durability in glasses containing high levels of aluminum is nepheline (NaAlSiO4) crystallization, as it is generally accompanied by a measurable decrease in the glass's chemical durability. This work seeks to understand nepheline crystallization, within the context of thermal history, and to elucidate the influence of compositional shifts in the residual glass (after crystallization) on the measured durability. The results presented within show a distinct deviation in leaching behavior as a function of structural makeup (calculated Q units). This understanding will provide practical information required for broadening glass compositional regions needed to more efficiently vitrify HLW.

Network Formation in Borosilicate Glasses with Aluminum or Gallium: Implications for Nepheline Crystallization

“…Its crystallization in HLW-immobilized glasses is to be avoided, as nepheline-bearing glasses exhibit reduced chemical durability potentially allowing the radioactive isotopes incorporated in its structure to leach out. While the undesirable effects of nepheline crystallization on glass durability can be limited to acceptable levels by controlling the residual glass composition and forcing crystallization to the glass surface, preventing crystallization altogether would be advantageous. The addition of B 2 O 3 to aluminosilicate glasses is known to inhibit nepheline crystallization, − and various reasons for this have been proposed.…”

“…Upon adding high fractions of Al to silicate glasses, the phase separation of crystalline nepheline is often observed, 21−29 an aluminosilicate which incorporates alkali, alkaline-earth, and transition metals 23 into its structure. Although aluminosilicates are generally chemically durable, 30,31 devitrified nephelinebearing glasses are not, 17,32 as they exhibit faster dissolution kinetics, whereby contained radionuclides can leach into the environment. The addition of boron to aluminosilicate glasses suppresses nepheline crystallization.…”

“…Upon adding high fractions of Al to silicate glasses, the phase separation of crystalline nepheline is often observed, − an aluminosilicate which incorporates alkali, alkaline-earth, and transition metals into its structure. Although aluminosilicates are generally chemically durable, , devitrified nepheline-bearing glasses are not, , as they exhibit faster dissolution kinetics, whereby contained radionuclides can leach into the environment. The addition of boron to aluminosilicate glasses suppresses nepheline crystallization. − Reasons advanced to explain this effect include an increase in the coordination number of Na + upon substituting Al by B, competition for Na + between [BO 4 ] − and [AlO 4 ] − units, noncompliance with the aluminum avoidance rule, improvement in the thermal stability, and an increase in the liquidus viscosity due to the substitution of Si–O–Al by Si–O–B linkages .…”

Structural dependence of crystallization in phosphorus‐containing sodium aluminoborosilicate glasses

“…A number of studies have found ND constraint to be conservative in the sense that there exists a composition space where glasses with N Si < 0.62 do not form nepheline. 1,2,[5][6][7][8][9][10][11] For example, the presence of boron has been shown to suppress nepheline formation. 3,8 Some studies have deliberately varied CaO and B 2 O 3 levels in waste glass formulations in order to understand these phenomena.…”

Predicting nepheline precipitation in waste glasses using ternary submixture model and machine learning

“…Most recently, a new descriptor, "difference based on correlation", was introduced by Sargin et al, 8 which has a higher accuracy compared to previous (single) descriptors, and also has more balanced false positive and false negative rates. In a study by McClane et al, 9 influence of compositional shifts in the residual glass on the measured durability was studied and a distinct deviation in leaching behavior as a function of a structural feature (Q unit) was observed. In addition to these efforts, machine learning (ML) approaches such as artificial neural network, 2 support vector machines, random forests, and decision trees 8 were also investigated to test their ability to predict nepheline precipitation.…”