where V is the liquid kinematic viscosity and Q is the gas volumetric flow rate. This equation is dimensionally consistent. The relative effect of liquid viscosity can be obtained by comparing the bubble diameters calculated from Eqs. (14-206) and (14-208). If liquid viscosity appears significant, one might want to use the long and tedious method developed by Krishnamurthi et al. (op. cit.) and the review by Kulkarni et al. (op. cit.) that considers both surface-tension forces and viscous-drag forces.

Intermediate Regime This regime extends approximately from a Reynolds number of 200 to one of 2100. As the gas flow through a

FIG. 14-92 Bubble-diameter correlation for air sparged into relatively inviscid liquids. Db = bubble diameter, D = orifice diameter, Vo = gas velocity through sparging orifice, P = fluid density, and ^ = fluid viscosity. [From Can. J. Chem. Eng., 54, 503 (1976).]

submerged orifice increases beyond the limit of the single-bubble regime, the frequency of bubble formation increases more slowly, and the bubbles begin to grow in size. Between the two regimes there may indeed be a range ofgas rates over which the bubble size decreases with increasing rate, owing to the establishment of liquid currents that nip the bubbles off prematurely. The net result can be the occurrence of a minimum bubble diameter at some particular gas rate [Mater, U.S. Bur. Mines Bull. 260 (1927) and Bikerman, op. cit., p. 4]. At the upper portion of this region, the frequency becomes very nearly constant with respect to gas rate, and the bubble size correspondingly increases with gas rate. The bubble size is affected primarily by (1) orifice diameter, (2) liquid-inertia effects, (3) liquid viscosity, (4) liquid density, and (5) the relationship between the constancy of gas flow and the constancy of pressure at the orifice.

Kumar et al. have done extensive experimental and theoretical work reported in Ind. Eng. Chem. Fundam., 7, 549 (1968); Chem. Eng. Sci, 24, part 1, 731; part 2, 749; part 3, 1711 (1969) and summarized in Adv. Chem. Eng., 8, 255 (1970). They, along with other investigators—Swope [Can. J Chem. Eng., 44, 169 (1972)], Tsuge and Hibino [J. Chem. Eng. Japan, 11, 307 (1972)], Pinczewski [Chem. Eng. Sci., 36, 405 (1981)], Tsuge and Hibino [Int. Chem. Eng., 21, 66 (1981)], and Takahashi and Miyahara [ibid., p. 224]— have solved the equations resulting from a force balance on the forming bubble, taking into account buoyancy, surface tension, inertia, and viscous-drag forces for both conditions of constant flow through the orifice and constant pressure in the gas chamber. The design method is complex and tedious and involves the solution of algebraic and differential equations. Although Mersmann [Ger Chem. Eng., 1, 1 (1978)] claims that the results of Kumar et al. (loc. cit.) well fit experimental data, Lanauze and Harn [Chem. Eng. Sci., 29, 1663 (1974)] claim differently:

Further, it has been shown that the mathematical formulation of Kumar's model, including the condition of detachment, could not adequately describe the experimental situation—Kumar's model has several fundamental weaknesses, the computational simplicity being achieved at the expense of physical reality.

In lieu of careful independent checks of predictive accuracy, the results of the comprehensive theoretical work will not be presented here. Simpler, more easily understood predictive methods, for certain important limiting cases, will be presented. As a check on the accuracy of these simpler methods, it will perhaps be prudent to calculate the bubble diameter from the graphical representation by Mersmann (loc. cit.) of the results of Kumar et al. (loc. cit.) and the review by Kulkarni et al. (op. cit.)

For conditions approaching constant flow through the orifice, a relationship derived by equating the buoyant force to the inertia force of the liquid [Davidson et al., Trans. Instn. Chem. Engrs., 38, 335 (1960)] (dimensionally consistent),

0 0

Post a comment