A systematic model-based approach is used for development of an
efficient carousel ion-exchange
process for the selective removal of radioactive
137Cs+ from alkaline nuclear waste
solutions.
Equilibrium data for two resorcinol−formaldehyde (R−F)
cation-exchange resins are correlated
by an empirical equation of the Freundlich−Langmuir type over
cesium/sodium concentration
ratios of 10-9 to
10-2 and sodium concentrations of 1 to 6 N.
The standard deviations are 3.5
and 6.6%, respectively. The data cannot be accurately described
by mass action equations. A
detailed rate model, developed in this study for the periodic
countercurrent multicolumn operation
of carousel systems, is used with the equilibrium correlations to
simulate cesium breakthrough
curves from R−F resin columns. Results show that accuracy of the
predicted breakthrough
curves are directly related to the accuracy of the isotherm data and
correlations. Cesium
breakthrough position is generally predicted to within 5% or less for
10 of 13 runs over linear
superficial velocities of 0.16 to 8.8 cm/min, column lengths of 3.14 to
118.5 cm, and particle
radii of 145 to 200 μm. One run shows later breakthrough than
predicted as a result of a low
potassium concentration in the feed. Two other runs show early
breakthroughs as a result of
channeling in poorly packed columns of a carousel system. Despite
the channeling, strong
thermodynamic self-sharpening effects helped establish constant pattern
waves in the downstream columns. A case study for a pilot-scale carousel unit shows
that 100% utilization of
cesium capacity and maximum throughput can be achieved while containing
the mass transfer
zone within the downstream columns. Since intraparticle diffusion
controls spreading of the
breakthrough curves, reducing the particle radius from 200 to 145 μm
increases throughput by
40%.