Cross-Flow Filtration Performance During the Washing of a Simulated Radioactive Waste Stream

Bechtel National, Inc. has been contracted by the Department of Energy to design a Waste Treatment and Immobilization Plant (WTP) to stabilize liquid radioactive waste that is stored at the Hanford Site as part of the River Protection Project (RPP). Because of its experience with radioactive waste stabilization, the Savannah River National Laboratory (SRNL) of the Westinghouse Savannah River Company is working with Bechtel and Washington Group International, to help design and test certain parts of the waste treatment facility. One part of the process is the separation of radioactive solids from the liquid wastes by cross-flow ultrafiltration. To test this process a cross-flow filter was used that was prototypic in porosity, length, and diameter, along with a simulated radioactive waste slurry, made to prototypically represent the chemical and physical characteristics of a Hanford waste in tank 241-AY-102/C-106. To mimic the filtration process the waste slurry undergoes several steps, including dewatering and washing. During dewatering the concentration of undissolved solids (UDS) of the simulated AY102/C106 waste is increased from 12 wt% to at least 20 wt%. Once at the higher concentration the waste must be washed to prepare for its eventual receipt in a High Level Radioactive Waste Melter to be vitrified. This paper describes the process of washing and filtering a batch of concentrated simulated waste in two cycles, which each containing 22 washing steps that used approximately 7.7 liters of a solution of 0.01 M NaOH per step. This will be the method used by the full-scale WTP to prepare the waste for vitrification. The first washing cycle started with the simulated waste that had a solids concentration of 20 wt% UDS. This cycle began with a permeate filter flux of 0.015 gpm/ft² (3.68 cm/hr) at 19.6 wt% UDS with a density of 1.33 kg/L, consistency of 19.1 mPa·s, and yield stress of 8.5 Pa. At the end of the 22 washing steps the permeate filter flux increased to 0.023 gpm/ft² (5.64 cm/hr) at 20.1 wt% UDS with a density of 1.17 kg/L, consistency of 12.6 mPa·s, and yield stress of 10.4 Pa. The average permeate filter flux during the 7 hours of Cycle 1 washing was 0.018 gpm/ft² (4.41 cm/hr). During Cycle 2 the simulated waste started at a permeate filter flux of 0.025 gpm/ft² (6.13 cm/hr). Note that the starting flux for Cycle 2 was greater than the ending flux for Cycle 1. The period between the cycles was approximately 12 hours. While no filtering occurred during that period either solids dissolution continued and/or the filter cake was dislodged somewhat with the stopping and starting of filter operation. At the end of the second set of 22 washing steps, the permeate filter flux increased to 0.032 gpm/ft² (7.84 cm/hr) at 20.6 wt% UDS with a density of 1.16 kg/L, consistency of 9.0 mPa·s, and yield stress of 8.2 Pa. The average permeate filter flux during the 4 hours of Cycle 2 washing was 029 gpm/ft² (7.11 cm/hr).