(301b) Modeling Hanford-Rpp Treated Law Feed Evaporation

Previous Hanford-RPP evaporator modeling has focused on the treated LAW feed and eluate evaporation systems without the inclusion of secondary waste recycles. This work investigates the potential impact that secondary-waste recycle streams may have on the operation of the treated LAW evaporator. The treated LAW evaporator will concentrate the treated waste effluent streams from the Cs ion exchange blended with the LAW melter offgas scrubbing recycle stream. The Tc ion exchange system (originally part of the test specification) has since been eliminated from the flowsheet. The LAW melter offgas scrubbing recycle stream is the major contributor to the overall recycle volume that is to be mixed with the treated waste feed prior to evaporation. Based on experience at Savannah River Site, the introduction of silica from melter offgas recycle into high sodium/aluminum feeds can produce sodium aluminum-silicate precipitates upon concentration. These sodium-aluminum-silicates can cause operational shutdowns due to plugging of lines and fouling of heat transfer surfaces.

This work examines the potential of the treated waste feed blends to form sodium-aluminumsilicate precipitates when evaporated using the zeolite database. To investigate the behavior of the blended pretreated waste feed, an OLI ESP model of the treated LAW evaporator was built. A range of waste feed compositions representative of Envelope A, B, and C were then fed into the OLI model to predict various physical and chemical properties of the evaporator concentrates. Additional runs with treated LAW evaporator were performed to compare chemical and physical property model predictions and experimental results for small scale radioactive tests (S-69) of the treated feed (AW 101) evaporation process.

The objectives of this work were to develop physical property correlations of the concentrated treated feed evaporator bottoms. The model was to simulate the treated LAW evaporator operating at 50°C at the bubble point vacuum with Envelope A, B, or C wastes blended with LAW melter offgas scrubbing recycle (SBS) as feed and the evaporator concentrate or bottoms stream being varied between 15°C and 66°C. The physical property correlations were to be expressed in terms of the waste feed compositions, LAW SBS to waste feed volumetric flow ratio, and the evaporator bottoms temperature and sodium concentration. Simulation results were then regressed to generate predictive equations for density, heat capacity, thermal conductivity, viscosity, and solubility of the Treated LAW evaporator bottoms concentrate. Simulation results were validated with experimental results which had already been completed.

Development of these equations was successful based on the goal of developing physical property correlations for each waste envelope with an error no greater than ±15% between calculated and modeled physical properties. The equation to predict solubility or the amount of total solids in the Treated LAW evaporator bottoms concentrate could not be developed to satisfy this goal. There was not enough information captured by the chosen model inputs to adequately describe the complicated nature of solids precipitation in the bottoms concentrate. However, general trends in solids formation were identified with respect to the bottoms Na molarity and temperature.

Another objective of this task was to verify the derived physical property equations with experimental tests with simulated blended LAW feed solutions. The predicted physical properties were compared with experimental results documented in the report Treated LAW Feed Evaporation: Physical Properties and Solubility Determination and with radioactive experiments for AW-101 documented in the report Evaporation of Pretreated Hanford Tank AW-101 Sample Mixed with Recycle. The predicted densities and heat capacities were within +/-15% of the measured values for Envelopes A, B, and C. The Na molarity predictions for Envelope A and B were also within +/-15%. The other predicted physical properties were outside +/-15% of the measured values. This mismatch with measured values was due in part to being outside the range of the predictions, comparing a measured slurry property with a predicted supernate property, and the exclusion of solids in the predictions.