The factors affecting the efficacy of a closely packed bed of mixed cotton and a supporting fiber (Teflon, glass, Dynel) were evaluated for oil-in-water dispersions. Several other waterorganic systems were also tested. Superficial velocities ranged from 0.2 to 3.5 ft./min. Succeiful coalescence was attained a t interfacial tensions as low as 3.5 dynes/cm. Dispersed phase viscosity was varied from 1.4 to 137 centipoise. For a mixed-fiber bed with a specific ratio of fiber species, there is an optimum bed depth for best performance; High-speed cinephotomicrographic observations a t 1 0 0~ and up to 4,000 frames/sec. indicated that fiber wettability is not the most important factor for successful operation.Organic chemicals must be purified during their manufacture by contacting them with immiscible solutions in a washin or extraction operation. The immiscible solution the field of petroleum products such as gasoline and kerosene. Fuel for modem jet-powered airplanes is a articularly important product in which the presence of water is a very critical problem (7). The reverse situation, in which very fine droplets of an oil phase are dispersed in a large amount of water, is of considerable interest in the prevention of contamination of rivers, harbors, and beaches.If a water phase is dispersed in an organic liquid by any of the efficient turbulence creating mixers presently in use, an emulsion will be f o n e d . If both liquids are pure (no surfactants present), the emulsion will be of a temporary nature and both phases may be recovered by a simple settling operation. If a strong surfactant is present, the emulsion may be a permanent one requiring special methods to break it. Two types of temporary emulsions are recognized (1 2 ) . The first (primary) is characterized by a drop size of the order of 100 p and will separate readily in a few seconds or minutes by simple settling. If one or both of the two liquids involved contains polar compounds, one or both of the two layers which result from the settling operation will be clouded with a fine mist or fog of extremely fine droplets of the opposite phase (1). Such fogged layers are called secondary emulsions and consist of untold billions of droplets of submicron size suspended in a fieId of the opposite phase. These secondary emulsions cannot be settled clear for many minutes or hours. Coalescence of these tiny droplets into large ones is necessary if the emulsion is to be broken and separation attained.Passage of an emulsion through a fibrous bed will often cause coalescence and facilitate separation (1 2) ; sometimes such beds are dependable and sometimes not. The mechanism of their operation is shrouded in vagueness and conjecture. No accepted theory exists on how they accomplish the coalescence of the submicron droplets into large ones of manageable size. Were one available, it would facilitate design and aid selection of materials of construction of such devices. The present work is con-

No abstract

radius of tube in calculations for reaction along cylindrical channel. cm dimensionless reaction rate, jR&(plate) ; {% (cylinder) average reaction rate over total reactive surface Reynolds number, dvp/g seconds average velocity, centimeter/second dimensionless velocity profile, for laminar flow, distance in axial direction, cm distance in transverse direction, cm3h20 plate) 1 = dimensionless position along y coordinate, y/h p = density, grams/cm3 literature Cited Acrivos, A,, ChambrB, P. L., I n d . Eng. Chem.49,1025 (1957). Pn = apparent reaction order for weak acids = coefficients in series approximation t o boundary layer Droblem y r ( n ) = gamma function of variable n, tabulated function = dimensionless radial position, r/R A pore model for the flow of a single-phase fluid through a bed of random fibers is proposed. An effective pore number, Ne, accounts for the influence of dead space on flow; deflection number, N6, characterizes the effect of fiber deflection on pressure drop. Experimental data were obtained with glass, nylon, and Dacron fibers of 8to 28-micron diameter and with fluids of viscosity ranging from 1 to 22 cp. A generalized friction factor-Reynolds number equation i s presented. The effects of dead space in a fibrous bed on flow and of fiber deflection on pressure drop have no parallels in a granular bed.

A mass transfer model for vigorously oscillating single liquid drops moving in a liquid field has been developed with the concepts of interfacial stretch and internal droplet mixing. The model takes into account both amplitude and frequency of drop oscillations. Experimental values of fraction extracted were predicted with an average deviation of 15%. Oscillations break up internal circulation streamlines and a type of turbulent internal mixing is achieved.The contact of large drops of a dispersed liquid phase with a continuous liquid phase roceeds in fain distinct fect the extraction operation. These four major zones for mass transfer in the lifetime of such Iarge drops are: formation of the discrete drops while they are still resident on the drop-forming device, an acceleration period immediately after the drops leave the nozzle or orifice, during the free rise (or fall) of the drops at a steady state slip velocity, and in a zone of flocculation and coalescence at the end of their vertical travel through the equipment (or stage) under consideration.We are concerned here only with drops in the freely falling zone and of such a size that they will exhibit a cyclic, oscillatory motion as they move through the continuous phase.The literature dealing with mass and heat transfer between fluid particles and their fluid surroundings is very extensive (9, 15, 18, 19). Most studies have tried to isolate the resistances to such transfer into an internal and an external one, relative to the phase interface. The resistance to transfer, whether internal or external to the droplet surface, depends upon the motion of the fluid particle. Widely ranging magnitudes of resistance have been reported for drops which behave as rigid spheres, those performing like nonoscillating fluid ( circulating) bodies, and those exhibiting a fully oscillating regime. Oscillating drops show a far greater rate of transfer than any other type. Summaries of the published work on the continuous (external) phase resistance are readily available (4, 13, 17, 18). Garner and Tayeban (3) studied the effect that droplet oscillation had on the continuous phase resistance.The dispersed phase resistance has been analyzed on the basis of three mathematical models. One of these, developed by Newman (14), is based upon a rigid sphere with no internal motion and leads to the expression stages if a spray or perforated p P ate tower is used to efWhen resistance to transfer in the continuous phase is zero and laminar circulation patterns which can be described by the Hadamard ( 5 ) streamlines are present, the result of Kronig and Brink (10) can be applied:Handlos and Baron (6) superimposed a turbulence due to random radial motion upon a circulatory pattern; this model yields, for zero resistance in the continuous phaseThe numerical values of An and 1 , in Equations (2) and ( 3 ) are not identical. Previous workers (1, 3, 8, 12, 19) have used these three theoretical models, combined with one of the various empirical equations for the continuous phase resistance, to...

A model is presented to account for reduced mass transfer to drops falling through a continuous phase which contains a surface active agent. The fluid flow patterns are essentially laminar. The reduction in mass transfer is said to be due to a reduction in available interfacial transfer area and to changes in both velocity and pattern of internal circulation. These are shown to be functions of contact time and can be characterized. Experimental values agreed with the theoretically predicted ones with a deviation of less than 10%.