The Hanford Site, situated in Washington State, houses 56 million gallons of radioactive waste historically stored in 177 underground tanks. To mitigate the risk posed by this hazardous material, the United States Department of Energy is finishing the construction of the Hanford Waste Treatment and Immobilization Plant (WTP). The WTP will utilize Joule-heated ceramic melters to vitrify the low-activity waste (LAW) and high-level waste (HLW) into borosilicate glass at a temperature of 1150°C. Ensuring the integrity of the melting vessel is a critical technical issue – its failure during vitrification can result in severe safety and financial consequences. Thus, such melting vessels require specific refractory liners to withstand high temperatures and corrosive environments for an extended period.
The Monofrax K-3 refractory has been identified as the optimal refractory material to be used in the melters at the Hanford site. K-3 refractory is a ceramic material that consists of a combination of (Mg, Fe)O·(Al, Cr)2O3-based spinels, corundum-based solid solutions, such as (Al, Cr)2O3, and a small quantity of SiO2. Its corrosion resistance is attributed to the elevated Cr2O3 content (about 27 wt. %) that exhibits low solubility in borosilicate-based glass melts. The high-temperature and corrosive environments of the vitrification process impose both chemical and physical wear on the refractories. The dissolution of the refractory material and the destruction of intergranular bonding due to chemical interactions between the K-3 refractory and the melt lead to the material’s degradation. Additionally, erosion causes removal of the dispersed refractory grains.
In the past, a number of studies evaluated the K-3 corrosion as a function of glass composition, developing empirical property-composition models to predict the compositional dependence of K-3 refractory corrosion (neck loss) during the vitrification of LAW glasses. However, much less attention has been paid to the effects of melt velocity. Thus, in this contribution, we will report initial results obtained from the dynamic corrosion testing. During dynamic corrosion tests, the refractory samples are eccentrically fixed in the platinum crucible (height 10 cm, width 10 cm) filled with molten glass. The velocity of the melt flow around the refractory material sample is regulated by the speed of rotation of the crucible with the glass melt. A range of flow velocities was tested based on the velocities obtained by the CFD simulations of the whole waste glass melter. After completion of each test, the refractory plates (coupons) were removed from the glass, cooled, embedded in epoxy resin, and sectioned. The corrosion rate in the suspended part and in the flux line region (neck loss) was measured by an image analyzer. The measured corrosion rate was evaluated as a function of melt flow velocity and temperature, and will be used to provide data for the development of the corrosion model to be implemented in the CFD model of the whole waste glass melter.