3.2. Effect of passes and pressures on final droplet size
In order to determine the effect of pressure and passes on the
formation of W/O emulsions a series of coarse emulsions were
made with 10 wt.% water containing 2 wt.% calcium chloride in
the aqueous phase and 9 wt.% PGPR in the oil continuous phase.
These emulsions were passed 5 times through the high-pressure
devices at 50 and 100 MPa (Fig. 2).
It can be seen from Fig. 2 that the homogenising efficiency of
both devices is similar and each pass continues to decrease the
emulsion droplet size for both pressures. This is in contrast to
the work done on the Microfluidizer with sunflower oil in water
emulsions where it was shown that the minimum droplet size
was observed after the first pass (Qian and McClements, 2011;
Lee and Norton, 2012; Lee et al., 2012). To understand the difference
the continuous phase viscosity needs to be considered. In
the oil continuous emulsion the continuous phase viscosity is higher
resulting in greater energy dissipation, but as a result the majority
of the energy dissipates prior to the jets impinging leading to
reduced shear in the impinging plane, and less efficient break-up.
Droplet break-up is similar in both devices suggesting that the
droplet break-up is less dependent on geometry as with O/W
homogenisation. As such this suggests there is little or no effect
from the impinging jets indicating that the majority of energy
dissipates before they impinge.
The effect of increasing the pressure from 50 to 100 MPa is
shown in Fig. 2, as expected there is a decrease in droplet size
for both devices as the higher pressure energy input creates larger
deformation stresses. However as can be seen for the HPH there is
an increase in droplet size after 5 passes. The HPH used in this
work had no temperature control during homogenisation, as a result
the emulsion temperature after five passes was approximately
30 C higher than the starting temperature, leading to an increased
rate of coalescence. Previous studies on water continuous high
pressure homogenisation have shown that in-processing coalescence
is present (Jafari et al., 2008; Lee and Norton, 2012). This
can be checked by comparing the overall droplet size distributions
(Fig. 3).
Fig. 3 shows that for both devices after one pass the droplet size
distribution is mono-modal and has no shoulders. This would indicate
signs of coalescence or flow bypassing the main turbulent
region therefore it is hypothesised that there is minimal inprocessing
coalescence. An explanation for the decrease in droplet
size with passes thus lies in the timescales of droplet break-up. The
time required to create droplet deformation and break-up must be
similar to the residence time of the stresses that create this droplet
deformation therefore multiple passes are required.