PNNL

Abstract

During the vitrification of nuclear waste, significant amounts of hazardous emissions are generated from the feed-to-glass conversion reactions, in addition to the flows from forced air bubbling, and air in-leakage. Although the major gaseous emissions are H2O, N2, O2, and CO2, various monitored environmental pollutants are also released, such as NO, N2O, NO2, CO, or SO2. Moreover, as by-products of reactions between organics and nitrates in the feed, the melter off-gas can also include other trace emissions with potentially hazardous impact, such as acetonitrile (CH3CN), acrylonitrile (C3H3N), and ammonia (NH3). These emissions are of concern for the Hanford Waste Treatment and Immobilization Plant, which plans to add sucrose to the feed as a reducing agent. Although off-gas emissions have been measured during laboratory- and pilot-scale melter tests, no predictive tool is currently available to estimate the composition of gaseous emissions for the hundreds of unique waste feed compositions that will be vitrified at Hanford. In this work, we address this issue by measuring the gas evolution for a broad range of low-activity waste melter feeds using laboratory evolved gas analysis (EGA), and developing a relationship between the feed composition and the off-gas composition. We demonstrate that next to the content of nitrogen and organic carbon in the feed, the gaseous emissions are strongly affected by the redox of the feed. We also show that while the gas evolution can be estimated directly from the feed composition using regression analysis, significant differences in gaseous emissions can exist between a laboratory EGA setup and pilot-scale melter operation. The results presented in this work provide a tool that can significantly reduce the amount of expensive pilot-melter testing and enable rapid optimization of feed composition with respect to off-gas composition.