Flow maldistribution in packed beds: A comparison of measurements with predictions

Abstract

Experimental measurements are reported on flow maldistribution in packed beds containing side streams and deliberately created spatially nonuniform resistance to flow. The actual experimental technique involved the determination of the velocity field of the gas stream exiting the column through the use of a hot wire anemometer. The experimental measurements were compared with predictions based on the numerical solution of the differential vectorial form of the Ergun equation and reasonable agreement was obtained. Both the measurements and the analysis confirmed the existence of preferential flow in the vicinity of the walls even for uniformly packed beds.

Effect of Interface Resistance on Gas Flow in Blast Furnace

“…As expected, the larger the disparity in sizes, the lower will be the numerical values of these minima. These results were used by different research groups [5][6][7] as a first approximation with qualitatively similar results. In blast furnace burdens this factor will be significant since there are many layers and hence many interfaces.…”

Mass Transfer in Packed Bed Electrochemical Cells Having Both Uniform and Mixed Particle Sizes

“…Values of jD calculated by Eq. [7], however, are within 3% and 1% of the values calculated with use of constants in Eq. [8] and [9], respectively.…”

“…[7] is a simplification. Since limiting current data were corrected by subtraction of background current, the difference in slope would not be attributed to a second electrolysis reaction.…”

“…1 illustrate how Eq. [7] and [10] correlate data in nonuniform beds. The lower portion of part A contains examples of results for mixtures of two bearings.…”

A new pseudo-continuous model for the fluid flow in packed beds

“…This corresponds to a reactor of considerably more than 7-8 particle diameters length, after which the pressure gradient has been observed to become constant along the axis (van der Merwe and Gauvin, 1971; Rowe and Henwood, 1961). Under this condition the pressure is also constant over the cross-section (Bi and Grace, 1996;Sparrow et al, 1976;Szekely and Poveromo, 1975), so that the pseudo-continuous flow can be assumed to be parallel to the reactor axis everywhere. This simplifies the problem considerably, resulting in a single ordinary differential equation for the radial distribution of the axial velocity component, where the pressure gradient is obtained from integral continuity (Eisfeld, 1999).…”