Evaluation of 5-cm Centrifugal Contactor Hydraulic and Mass Transfer Performance for Caustic-Side Solvent Extraction of Cesium

A test program has been conducted in which the use of pilot-scale centrifugal solvent extraction contactors for cesium removal from an alkaline waste stream has been successfully demonstrated. The program was designed specifically to evaluate the use of centrifugal contactors having 5-cm-diam rotors for the removal of cesium from alkaline high-level waste (HLW) that was generated and is being stored at the U.S. Department of Energy's Savannah River Site (SRS). The removal of cesium from this waste is highly desirable because it will reduce the volume of waste that must be treated and disposed of as HLW. The parameters applied in the test effort are those that have been established for the Caustic-Side Solvent Extraction (CSSX) process, a multistage extraction operation that has been designed by researchers at Oak Ridge National Laboratory (ORNL) and Argonne National Laboratory (ANL). In the CSSX process, cesium is extracted by calix(4)arene-bis-(fert-octylbenzo-crown-6), commonly referred to as BOBCalixC6. The extract is scrubbed with dilute (0.05 M) nitric acid, both to remove coextracted elements (primarily potassium and sodium) and to adjust the pH of the extract to facilitate recovery of the cesium. The scrubbed solvent is contacted with 0.001 M HNO{sub 3}, which results in the stripping of the cesium from the solvent into the aqueous acid. The CSSX process flow rates have been established so to produce a cesium concentration in the strip effluent that is 12 to 15 times the concentration in the waste stream that enters the extraction section of the cascade. Results from initial hydraulic testing of a commercially available 5-cm contactor under CSSX conditions indicated that the mixing of feed solutions within the unit (which is critical to efficient solute transfer) was limited by a feature of the contactor that was designed to increase throughput and improve separation performance. In the design, phase separation is improved by reducing turbulence within the contactor. Subsequent to the initial hydraulic test: cesium transfer tests were performed using contactors arranged in both single-stage and multistage arrangements. Results of these tests confirmed that phase mixing within the contactor was inadequate. In an effort to improve mixing within the contactor and thereby increase mass transfer efficiency, two minor modifications were made to a single contactor unit. One modification was the replacement of the bottom plate from the vendor-supplied contactor housing, which was equipped with curved (impeller-type) vanes, with a bottom assembly that had straight radial vanes. The latter configuration is the standard used in all existing ANL, ORNL, and SRS contactor designs. The second modification involved enlargement of the opening in the bottom of the rotor through which dispersion from the contactor mixing zone enters the rotor for separation. By increasing the rotor opening sufficiently, the rotor loses pumping efficiency to such an extent that accumulation of a hydrostatic head in the annular mixing zone is required for solution to be pumped through the contactor to the organic and aqueous discharge ports. By causing a volume of liquid to accumulate in the mixing zone, it is expected that phase mixing will be improved. Following modification of a contactor, hydraulic testing was repeated to determine flow parameters to be applied in mass transfer testing using the modified device. As expected, test results indicated that the maximum throughputs that could be achieved using the modified contactor under extraction and stripping conditions were lower than those obtained using the ''as-received'' unit. However, phase separation performance within the reduced operating envelope was excellent. Most importantly, cesium transfer stage efficiencies were significantly improved over those obtained using the unmodified device and resulted in attainment of the target CSSX process decontamination factor of 40,000 when extrapolated to the baseline CSSX contactor cascade