In the ideal case the particles in the clump have a higher density than the liquid, all the gas in the interstitial space escapes before the clump drops below the surface, and after the particle sinks the region close to the upper surface of the liquid is simply clear liquid -- an optimum condition for wetting-in additional powder.

In most commercial processes the situation is not so simple. Figure 4-1 shows the behaviour of clunps for several problem situations.

Figure 4-1:
a) dense particles, b) light particles, c) non-wetting particles, d) foam

Particles much more dense than the liquid may create a clumps density that is high enough that the clump submerges before all the gas has been displaced. This trapped gas interferes with full wetting of the clump, may delay the breakup under shear, and may lead to bubbles in the final product (not good for paint or microcircuit paste applications).

Particles much less dense than the liquid may keep the clump density so light that the clump will stay afloat even when fully wet. If more dry powder is added it will pile up on the floating mass and not be wetted as efficiently as if it were falling on clear liquid.

Particles that are not wet well by the liquid may not gain enough density to become submerged, so it forms a floating mass containing air and if more dry powder is added it will pile up on the floating mass and not become wet at all.

Excessive agitation may result in air being drawn beneath the surface and then broken into small bubbles by the impellor. In the presence of a foam stabilizer this foam may persist as a bed on top of the liquid, so if more dry powder is added it will tend to pile up on the foam, become partially wetted, roll about to form dispersion-resistant granules, and thus not wet in well.

What lessons can we learn from this?

Powder should be dropped onto the surface of the liquid rather than being added subsurface or else added to the liquid under vacuum so that there is no air to displace. If gas remains in the clump it may lead to "microbubbles" in paint formulations or support the formation of foam.

Powder should be dropped into the tank near the walls where fluid upwelling occurs so that the clumps will have maximum time available to wet-in and submerge before the surface element is forcibly submerged by being drawn into the downward spiraling vortex. The same concept applies if powder is thrown at a wetted wall of liquid flowing down the inside of a pipe wall.

In an stirred tank the rate of powder addition should be slow enough to allow full submergence of the powder before the surface element returns to the the powder feed point. If the rate of addition is too high a raft of powder will build up and produce large balls of powder dry on the inside and wet on the outside. Capillary action will pull the liquid into the clump and the resulting forces will compact the powder into a compact mass -- usually very hard to deal with.

A rough guide to the effect of various system parameters on the time required for the liquid to be drawn into the clump by surface tension (displacing the gas from an unsubmerged clump) is given by the following equation (derived for an individual cylindrical pore).

Here d_pore is the pore diameter, which we may approximate by using the volume-average particle diameter, is the liquid viscosity, D_clump is the clump diameter (perhaps hard to estimate).

EXAMPLE: For wetting pigment into a paint vehicle, where the viscosity is about 100 times that of water, d_pore = 2 m, D_clump = 2 mm, = 0.1 Pa s, = 0, = 72.8 mJ/m² the estimated time for wetting is
t_wet = 4 * 0.1 * [2 x 10^-3]² / ([2 x 10^-6] * 0.0728 * 1)
    => 11 s.
This may help us understand why this plant is having problems at this wetin station. We might suggest that if they can breaking the clumps to 0.5 mm in diameter (just a bit better) the wetin time may drop by a factor of 16 and the problem may vanish.