(140c) Modeling of Saltcake Dissolution
AIChE Annual Meeting
2008 Annual Meeting
Nuclear Engineering Division
Chemical Engineering Advances in the Nuclear Fuel Cycle
Monday, November 17, 2008 - 4:05pm to 4:30pm
The Hanford site contains approximately 53 million gallons of waste stored in 177 underground storage tanks. A total of 149 single-shell tanks (SST's) and 28 double-shell tanks (DST's) are located in a number of tank farms. The SST's contain the bulk of the salt waste while the DST's predominately contain sludge and interstitial liquor. In order to more fully evaluate the extent of mixing during the salt cake dissolution process and to provide further data for model validation and flow sheet development, researchers at the Applied Research Center (ARC) at Florida International University (FIU) have performed pilot-scale experiments on simulant compositions derived from Hanford waste inventories for saltcakes in tanks S-112 and S-109.
The ability to predict the composition of waste streams arising from the dissolution of saltcake during retrieval operations is valuable in that such simulations of the dissolution process may allow for examination of various retrieval strategies prior to their use in the field. The Environmental Simulation Program, V7.0, with V7DBLSLT (the latest version of the double salt database developed by the Institute of Clean Energy Technology), was used to predict effluent stream compositions, residual saltcake composition, saltcake height for dissolution experiments. Based on the initial extent of pretreatment of the waste, the salt cake may be either unsaturated or hydraulically saturated. Using a stimulant based on S-109 saltcake, two dissolution scenarios were examined, one where the saltcake bed was saturated with brine, and one where the majority of the brine had been drained from the saltcake bed (unsaturated). The S-109 simulant was rich in sodium nitrate, but also contained sodium carbonate, trisodium phosphate, and sodium sulfate, along with minor contributions of fluoride, oxalate, nitrite and chloride. Model predictions were compared to experimental results from ARC. The comparisons demonstrate the utility of ESP to predict effluent stream compositions, effluent stream properties such as density, and estimated column saltcake height.