Back-mixing in liquid-liquid extraction spray columns

Backmixing of the continuous phase was studied in liquid-liquid spray columns of various geometries, for various flowrates of the two phases, and for various drop size distributions. The dispersion or eddy diffusion model was used to characterize the axial mixing of the continuous phase. Axial concentration profiles were measured upstream, with respect to the continuous phase, from a distributor of sodium chloride tracer (soluble in the continuous phase only). The steady state form of the model was utilized to calculate axial eddy diffusivities from these results. The tracer studies showed that the axial eddy diffusivity is independent of the continuous phase flowrate and the column height. Axial eddy diffusivities between 7-ft.²/hr. and 31-ft.²/hr. were obtained in a 1½-in. I. D. column. Low dispersed phase flowrates and large drop sizes resulted in high axial eddy diffusivities. Increasing the column diameter to 3-in. resulted in superficial axial eddy diffusivities between 6.3 and 17.3 times larger. The hold-up of dispersed phase was measured by means of a piston sampler. The hold-up increases approximately linearly with increasing dispersed phase superficial velocity and tends to be slightly higher for increased continuous phase superficial velocities. A smaller drop size resulted in an increased hold-up. Drop size distributions were measured. They always show two peaks, one at 0.02-in. diameter, and the other at a much larger size, the actual value of which depends on the nozzle tip diameter used to disperse the drops. The mixing cell-packed bed analogy was used to predict Peclet numbers in a spray column. The agreement between these and measured Peclet numbers is good for drops of about 0.15-in. equivalent diameter but becomes progressively worse as the drop size is reduced.Applied Science, Faculty ofChemical and Biological Engineering, Department ofGraduat