Shear Enhanced Transport in Oscillatory Liquid Membranes zyx David T. Leighton, Jr., Mark J. McCready z Department of Chemical Engineering z University of Notre Dame Notre Dame, IN 46556 The development of improved separation processes will play a major role in the successful commercialization of new products manufactured by genetically altered organisms, solving con- tinuing problems such as waste treatment and improving the overall competitiveness of the chemical industry. The objective of separation processes is to selectively remove a component from a mixture and concentrate it into a nearly pure form at rea- sonable rates, reliably and at an acceptable cost. While numer- ous separation schemes exist, lack of selectivity, low through- puts, long down times for recharging, and difficulties with han- dling thermally or chemically labile material are still limita- tions. As a consequence, recognizing their profound impact on the process industries, there is considerable room for improved separation schemes. One separation technique which has a great deal of unreal- ized potential is the use of liquid membranes. Noble and Way (1987 a) note that these membranes potentially provide both high selectivity and high solute flux through the use of chemical complexing agents which act as carriers, can be used for a reac- tion-separation device, and allow for the concentration of the product. In addition, cost of the liquid and carriers should not be a problem since relatively small amounts are used to impregnate the membranes and losses should be minimal. Economic analy- ses and pilot plant studies described by Noble and Way (1987 b) indicate that in its present state, liquid membrane technology should be competitive with liquid extraction for a number of processes. However, despite its promise, as of mid 1986 only one full-scale commercial venture used liquid membrane technology (Noble and Way, 1987 b). The lack of use of liquid membrane technology in the separa- tions industry may be ascribed largely to the difficulties of translating laboratory devices into large scale processes. Be- cause of costs and difficulties associated with creating and breaking stable emulsions, liquid surfactant membranes seem to be an attractive alternative to traditional processes only when species must be extracted to very low levels, such as recovery of metal ions or removal of organic compounds from water to be discharged into the environment. Supported liquid membranes avoid many of the problems of emulsions, however they suffer from difficulties associated with solvent stability within the membrane and from relatively low solute transport rates. The use of membranes of increased thickness improves solvent and membrane stability, but greatly reduces transport rates. Process reliability considerations further limit their use due to a need for short recharging intervals. In this paper, a novel procedure for dramatically increasing solute fluxes through supported liquid membranes, which should be readily realizable both in the laboratory and in process size equipment, is described. The imposition of a periodic pres- sure gradient transverse to the membrane is shown to increase the effective diffusivity of large molecules through the mem- brane by as much as two orders of magnitude or more, effec- tively eliminating the resistance to mass transport provided by the bulk of the liquid membranes. The enhancement provided by this periodic motion is superficially similar to that found in Tay- lor-Aris dispersion for steady flow in that the dispersion arises from the diffusion of the solute between streamlines with dif- fering velocities. However, the relative enhancement can be “tuned” by adjusting the frequency to an optimal value which should allow for a degree of selectivity, particularly in the con- centration of solutes from much smaller solvent molecules. In fact, as will be shown below, it is possible in principle to cause the effective diffusivity of a large molecule to actually be greater than that of a much smaller species. The dramatic increase in transport rates should allow for wider application of liquid membranes because the thinness of the support will no longer be the overriding consideration. Better membrane stability would be easily achieved by using a greater membrane thickness without significantly affecting transport rates, and the increased inventory of solvent + carrier should enable much longer operation times. In addition, if membranes which have a three-dimensional pore structure are used, it may AIChE Journal October 1988 Vol. 34, No. 10 1709