Application of Computational Fluid Dynamics in the Simulation of a Radioactive Waste Vitrification Facility

Abstract

The Columbia River in Washington State is at risk of radioactive contamination — a legacy of the cold war. Two hundred thousand cubic meters (fifty-three million US gallons) of radioactive waste is stored in 177 underground tanks at the Hanford site. This waste, which is 60% of the nation’s radioactive waste, is a product of 50 years of plutonium production for national defense. Bechtel National, Inc. has been commissioned by the U.S. Department of Energy to design and build a vast complex of waste treatment facilities to convert this waste into stable glass using a proven vitrification process. In this vitrification process, radioactive waste is mixed with glass-forming materials, then melted at approximately 1200C, and then poured into stainless steel canisters. These canisters are then permanently stored at secure aboveground or belowground facilities. The vitrification process results in a large amount of heat being stored in the hot glass. This heat must be removed within production schedule constraints. In the vitrification facility this glass is cooled in a small room called the Pour Cave. The room contains insulation to protect the concrete, and ventilation and water-cooled cooling panels to facilitate heat removal. The canister heat release rate depends on the thermal properties of the glass (which varies as the glass recipe changes), and the local environment, which includes other hot glass canisters. The cooling process is extremely complex. It is strongly coupled, and is driven by radiation, forced convection, natural convection and conduction heat transfer. Computational fluid dynamics, CFD, was used to predict the heat load to the ventilation system, the cooling panels and to the insulated concrete walls for a variety of operating conditions, providing the data needed for the design of these systems. Of particular interest was the temperature of the concrete, and whether or not design limits would be exceeded. The paper describes the special techniques that were developed to simulate the Pour Cave. This includes description of the modeling of the pouring of the glass, buoyancy modeling, and initialization of the simulation. Results are presented which show the predicted heat transfer characteristics throughout the Pour Cave.