Nuclear power generation leads to the production of a diverse and sizeable amount of nuclear waste. One of the largest challenges this waste presents is the radioactivity of fission products and by extension spent fuel. A very problematic elements in this waste is Cs-137 (caesium 137)which is highly radioactive, a high product yield of nuclear fission and also volatile which leads to difficulties when attempting to immobilise it through any thermal treatment method. This isotope is of concern in wastes directly as well as in the water of pools used to cool spent nuclear fuel.
As a result Cs-137 is regularly captured by the use of Cs (caesium) selective ion exchangers. This thesis concerns itself mainly with IONSIV IE-911 and clinoptilolite. Even though these ion exchangers are exceptionally good at selective removal of Cs from a variety of environments the, then spent, ion exchanger is usually in an unsuitable form to act itself as a direct disposal wasteform. This thesis considers the vitrification of these ion exchangers to create a consolidated, dense, chemically and radiation tolerant final wasteform. The ion exchangers considered have chemical compositions that relate to glass forming systems and therefore could be vitrified without large scale addition of glass forming additives. This thesis therefore investigates the creation and durability of a glass wasteform from loaded IONSIV IE-911 and the glass forming ability of the calcium aluminosilicate system (similar to clinoptilolite’s chemistry) when loaded with the alkali metals with a focus on Cs.
The work identified that glasses in the sodium titanosilicate phase diagram have the ability to retain Cs and their retention is in excess of 80% from batch. This retention level is found when IONSIV IE-911 is vitrified with small additions of soda and silica which indicates that the retention in this glass system is independent of processing methodology. However, before durability testing even began the most soda rich sodium titanosilicate glass displayed signs of poor durability in ambient conditions. A PCT-B (product consistency test B)dissolution test was conducted on a base glass, loaded base glass and IONSIV glass of the same target composition, the results found that indeed the titanosilicate glasses exhibited poor durability with large alteration layers and large normalised mass loss of constituent elements. The IONSIV glass, however, had normalised mass losses orders of magnitude lower than the basic sodium titanosilicate glasses and displayed no alteration layers after the 90oC 28 day test, indicating that a wasteform made from vitrifying IONSIV IE-911 would have good retention of the Cs it contains and be a more chemically durable wasteform than the equivalent glass made from laboratory chemicals; it is suggested that this is due to additional minor components present in IONSIV.
When investigating the calcium aluminosilicate glass system (which has a similar composition to clinoptilolite) a base glass series was first identified. However, once attempting to load this system with Cs the melts stopped forming. The failed melts were analysed and showed that they were mainly Cs aluminosilicates and calcium alumino/calcium silicates. It was unclear whether this was an effect exclusive to Cs, whether it would apply to all group one elements or is an effect seen as a trend throughout group one elements. As a result, a scoping melt was undertaken where base glasses were loaded with 1, 5 and 10 mol% Cs to attempt to identify a stoichiometry to continue with. The melt selected to continue on with the work was the 50 CaO 10 Al2O3 40 SiO2 (CAS 50 10 40) stoichiometry. This was loaded by cross elemental weighted substitution with 1, 5 and 10% of each of the alkalis in small alumina crucibles to synthesise multiple samples simultaneously. The Li, Na & K melts were easy to synthesise at 1500oC (apart from Li 5%, discussed later in this work), however, at 1500oC the Rb and Cs samples crystallised heavily at 5 and 10 mol% loading although the 5 mol% rubidium XRD indicated a more X-ray amorphous product than the 5% Cs sample. These were then re-synthesised at 1600oC at which temperature all the samples were X-ray amorphous bar 10 mol% Cs, indicating that the larger the alkali cation the more difficult it is to incorporate into the melt. The 10 mol% Cs sample was attempted at 1700oC to slightly more success and also at the same temperature in a zirconia crucible, still to slightly more success but there was still crystallisation. It has been assessed that due to the structural relationships in this glass system with the hybrid silica alumina network formation, cations have three potential roles to play in the structure as charge compensators on AlO4- – SiO4 bonds, on AlO4- - AlO4- bonds or associated to non-bonded oxygens (NBOs). The role of each cation is decided by its cation field strength which has been reported as affecting the coordination of alumina although this work does not confirm this. Therefore in order to incorporate Cs effectively into these melts it is imperative that the balance of Ca + Cs: Al is assessed and kept as much as possible to (Ca + Cs):Al = 1 in order to try and force all cations into network forming roles via charge compensation which will ultimately make a stronger and more durable wasteform.
