Loss of particles of different sizes from fluidized beds
IN THE course of some work on the loss of particles from fluidized beds, we have noted that the particles can be divided into three size ranges with respect to losses: particles whose Stokes Law terminal velocities are greater than the average velocity of the fluid in the free space, which are practically not lost at all; those whose Stokes Law terminal velocity is less than the minimum fluidization velocity of the bed, which are lost very rapidly; and intermediate sizes. To demonstrate this, we performed the following illustrative experiment.
The bed was made up of glass spheres. Two cuts were mixed: 3.00 weight per cent of spheres passing a 325-mesh standard sieve (i.e. less than 44 p dia.), and 97.00 weight per cent of a commercial size averaging about 100 p dia. The size distribution of each cut was determined microscopically (500 counts for the smaller size, 100 for the larger) and the number percentage distribution of the mixture computed.
The diameters were recorded to the nearest scale unit on an eye piece micrometer scale, checked against a Zeiss objective micrometer.
The mixture was put in a glass tube 30.5 mm in diameter to a depth of 5.0 cm, supported by a fritted glass disk, and fluidized with dry air at atmospheric temperature and pressure. Minimum fluidization velocity was determined by plotting pressure drop against air flow. The intersection of the two straight lines for the non-fluidized and the fluidized regimes was taken as the minimum fluidization velocity.
For the loss determination, the upper end of the glass tube was closed with a stopper containing a glass tube 50 mm long by 8 mm in diameter. From the bottom of this tube to the top of the bed was approximately 44 cm. The average velocity of the air in the 30.5 mm tube was IO.0 cm/set. Samples of the particles lost were obtained by holding microscope slides, made sticky by a thin layer of Vaseline, close to the exit tube. Two hundred oarticles were measured microscopically. The results were as shown in the table.
The minimum fluidization velocity was found to be 1.22 cm/set, which corresponds to a Stokes Law diameter of 13 b. The diameter corresponding to the average mean velocity was 37 flu. The table indicates that there is a sharp drop in the fraction of particles lost from the bed as the diameter increases from 16 to 18 p, or not much above the Stokes law diameter for the minimum fluidization velocity as we define it. Other methods of de6ning this velocity would give a larger value and better agreement with the break point. Also, if small particles preferentially escape collection on the slide, the effect would be to decrease the contrast between small and large particle losses, and perfect collection would show a greater effect.
There are substantially no losses above the Stokes law diameter for the average air velocity in the tube, although at this velocity, the Reynolds number is in the laminar flow range. This indicates that the time average velocity across the tube is approximately uniform just above the surface, as in a packed bed.