The point is of importance because of polydispersity:d 53 decreases during creaming,
and thereby Q, as is illustrated in Figure 13.11d. Assume, for example, that
Ro^2 =HT¼ 450 ?g=HV, that centrifuging time was 20 min, and that 90%creaming
was obtained. Hence 90%creaming would occur in 9000 min in the vessel, but it
often is more interesting to know when a little creaming would occur, say 3%of the
oil. Then it would be wrong to assume that it would take 3/90 as long, i.e., 5 hours; it
may be more like 1 hour if the emulsion is quite polydisperse (see also Figure 13.11d).
Finally, it should be checked that the droplet Reynolds number is not over 0.1 in the
centrifuge test, but that is rarely the case. (b) Assuming now that the conditions just
specified are met, the following may be causes for overestimating the creaming rate:
.Group sedimentation during centrifuging: see Figure 13.12. This will nearly
always occur, and it can be checked by calculating the Pe ́clet numbers in both
situations. The effect is not very large, rarely over a factor of 2.
.If the continuous liquid is strain rate–thinning, as illustrated in Figure 13.10
(curves 1 and 3), the effective viscosity may be much smaller in the centrifuge than
in gravity creaming, sincessedis byRo^2 /gtimes larger. For some systems, the
discrepancy may be large, up to a few orders of magnitude. It can be checked by
determiningZaas a function of shear stress.
.It may also be that the system has a yield stress (Figure 13.10, curve 2) and that
ssedis larger thansyin the centrifuge and smaller under gravity. That means fairly
rapid versus no sedimentation, respectively. The magnitude of a possible yield
stress can be determined in sensitive rheometers.
.A fairly trivial case is that the liquid in the vessel is subject to convection currents,
e.g., due to temperature fluctuations, whereas agitation is negligible in a centrifuge
(provided it has a swing-out rotor).
(c) If the emulsion droplets show slow aggregation, this may considerably enhance
sedimentation as soon as sizable aggregates have formed. However, this may take
some hours, and a centrifuge test can be done soon after the emulsion has been made
and can be completed in, say, half an hour. This then means that the test is not
suitable to detect aggregation-enhanced sedimentation. Aggregation can readily be
detected by microscopy.
13.4 COALESCENCE
Coalescence is induced by rupture of the thin film between close emulsion
droplets or gas bubbles; this phenomenon of film rupture will be discussed
first. However, the whole coalescence process involves a number of
additional variables, and these are rather different for emulsions and foams
(cf. Table 11.1), which is the reason that they are discussed separately. Our
understanding of coalescence is yet unsatisfactory, because (a) so many