Fig. 1 - Cross section of a typical Type IIIA SRP underground high-level radioactive waste tank, showing the injection of an inert (N2) purge jet, air inleakage (if purge jet is not operating), and benzene/hydrogen release from the liquid surface.
Flammable gases can be generated in DOE high-level waste tanks, including radiolytic hydrogen, and during cesium precipitation from salt solutions, benzene. Fig. 1 shows a typical underground high-level waste tank found at the Savannah River Site. Under normal operating conditions the potential for deflagration or detonation from these gases is precluded by purging and ventilation systems, which remove the flammable gases and maintain a well-mixed condition in the tanks. Upon failure of the ventilation system, due to seismic or other events, however, it has proven more difficult to make strong arguments for well-mixed conditions, due to the potential for density-induced stratification which can potentially sequester fuel or oxidizer at concentrations higher than average. This has complicated the task of defining the safety basis for tank operation, and in the case of cesium precipitation, has led to delays in cesium processing which are quite expensive. Improved tools for predicting tank mixing processes following loss of ventilation, coupled with mixing experiments designed specifically for DOE waste tank conditions, have the potential to both strengthen the safety basis for tank operation and to prevent schedule delays in tank operations.
Currently no numerical tools exist for modeling the transient evolution of fuel and oxygen concentrations in waste tanks following loss of ventilation, so highly conservative, bounding analysis methods must be used. When used with reasonable grid resolutions, standard multi-dimensional fluid dynamics codes suffer from excessive numerical diffusion effects, which strongly over predict mixing and provide nonconservative estimates, particularly after stratification occurs. The National Institute of Standards and Technology (NIST) has developed useful codes for predicting stratification and mixing due to fires in enclosures, but these codes are not supported by appropriate experiments for waste tanks, and do not consider mixing induced by injected jets, or the detailed distribution of fuel and oxygen concentration.
The UCB Thermal Hydralics Group is pursuing a concentrated effort to develop models and a numerical tool BMIX (Berkeley Mechanistic Mixing Model) to mechanistically predict mixing processes in large waste-tank volumes, where mixing processes can be driven by hot and cold vertical and horizontal surfaces and injected buoyant jets. We are supporting the model with scaled experiments using water/salt solutions and simulant fuels (helium and refrigerant-22 for hydrogen and benzene) to study the specific mixing processes which occur in waste tanks, and will also support the implementation of the code for use in waste tank operations.
We have developed the necessary scaling and analytical models to predict the transition from well-mixed to thermally stratified conditions in large enclosures, and have validated these models against both experimental data [A20] and high-resolution, multidimensional computational fluid dynamics (CFD) predictions [C24]. It appears that we can now describe, with good accuracy, mixing and transport processes in large complex enclosures, using simple coupled one-dimensional differential equations. Supported by specific experiments, these coupled equations can be solved with minimal computational expense.