Page 255 - Dust Explosions in the Process Industries

Generation of Explosible Dust Clouds 227

Fairchild, Tillery, and Wheat (1985) studied the reentrainment of fine, cohesive alu-
minum particles of <lo pm diameter in a wind tunnel, without and with large saltation
particles in the airflow sweeping over a fine-particle bed. The saltation particles were
monosized spheres of 100,240, and 500 pm diameter, and they were introducedinto the
airstream upstream of the bed of fine particles after stationary flow conditionshad been
established. Measurement of dust concentration as a function of distance above the bed
surfacewas conductedbetween 10 and 150mm. It was concluded that, within the exper-
imental range, resuspension of particles from a bed of loosely packed aluminum parti-
cles increased monotonicallywith increasing gas velocity and size of saltationparticles.
Singer et al. (1967) studied the entrainmentof coal and rock dust in an airstreampass-
ing over a loosely packed dust ridge placed on the floor of a laboratory scale wind tunnel,
as illustratedin Figure 3.22. The properties of the three dusts tested are given in Table 3.3

22mm
Figure 3.22 Cross section of typical dust ridge used in wind tunnel dust entrainment experiments.
The length of the ridge is 25 mm (From Singer et a/., 1967).

Table 3.3 Properties of dust used in dust ridge entrainment experiments

Dust type
Bulk density (g/cm3)
Solid density (g/cm3)
Porosity (E) 0.59 0.57
Median particle size (pm) 10 27
Source: Singer et al., 1967.

Photographicstudizesdisclosedvarious mechanismsof dust dispersion.These included
erosion from a dust surface and denudation from a dust surface under the influence of a
pulsating airstream.In erosion, the dust is dispersedparticle by particle from the deposit
surface. In denudation,the entire dust layer leaves the surface suddenly without the par-
ticles being separated at the instant of lifting. Denudation was considerably faster than
erosion for similar deposit geometries.
Even at air velocities only slightly higher than the minimum air velocity for particle
entrainment, the ridge dispersion was relatively rapid, having a characteristictime con-
stant of less than 0.1 s. Minimum air velocities for dust dispersion at half ridge height
above the wind tunnel floor were calculated to be 10-20 ds, using classical boundary
layer theory. There was no clear differencebetween the minimum velocitiesfor Pittsburgh
coal and the finer anthracite.However, as Table 3.3 shows, the finer anthracitehad a con-
siderably higher porosity than the coal; and this probably compensated for finer parti-
cles being more difficult to entrain than larger ones at the same bed porosity. As would
be expected, the bulk density of the dust ridge had a significantinfluenceon the minimum