(535g) On the Effect of Surfactants on Drop Coalescence with Liquid/Liquid Interfaces

1. Introduction

Drop
coalescence is commonly encountered in a wide range of industrial applications
in food processing and in petrochemical and pharmaceutical fields. In
such cases, the surface of the drop is often coated with surfactants, whose
properties largely affect the coalescence behaviour. In current oil extraction methods,
water is injected into the well to displace the residual oil in rock gaps and
enhance crude oil recovery. During injection, oil-water dispersions form. In
some cases, surfactants are also added to facilitate the formation of these
dispersions. Previous studies have shown that the oil recovery efficiency is
affected by this addition of surfactants. As it is difficult to trace the
distribution of the surfactants along the droplet surface, the effect of
surfactants on the coalescence process is still an active research area. In
this work, advanced laser velocity field measurements were conducted, using the
Particle Image Velocimetry (PIV) technique, to investigate the effect of
surfactants on the coalescence of single aqueous drop with an aqueous/organic interface.

2. Experimental Set-up

The
aqueous phase was 80% w/w glycerol/water mixture, while the organic phase was a
low viscosity kerosene-like Exxsol D80 oil. The concentration of the aqueous
mixture was chosen to match the refractive index of the organic phase,
necessary to minimize optical diffractions to produce more accurate laser based
flow measurements. Various concentrations of Span 80 were introduced to the organic
phase to study the effect of a surfactant on the coalescence mechanism. The
properties of the fluids are shown in Table 1.

Table 1:
Properties of fluids used.

The experiments were
conducted in a transparent acrylic tank of 50 mm square section width and 150
mm in height, as shown in Fig 1.
The aqueous/organic interface was placed at 40 mm from the bottom of the cell
while the organic/air was placed at 100 mm. Droplets of water/glycerol mixture
were generated by injecting the liquid into the Exxsol D80 through a
cylindrical stainless steel nozzle of 80 mm length and 2 mm inner diameter. The
nozzle was fixed at approximately 20 mm above the aqueous/organic interface to minimise
any variation on the droplet impact locations. For the investigation of the
velocity fields with PIV, the droplets were seeded with 1 micron Rhodamine tracer
particles. To clearly observe the interface between the aqueous and organic phase,
a small amount of Rhodamine 6G dye was introduced to the aqueous phase close to
the interface. These seeding particles were removed with a suction pump after
each experimental run.

Fig.
1.
Experimental set-up.

3. Results and
Discussion

In Fig. 2, the
variation of the relative height h_s of the droplet top point over
time t for various surfactant concentrations is plotted. Prior to the rupture
of the interface, the increasing surfactant concentration reduced the
interfacial tension and as a result the deformation of the interface was
increased. During the rupture process, for
the pure system or low surfactant concentration, φ
=0, φ
=2^.10^-5
and φ
=1^.10^-4,
the upper part of the droplet surface oscillated at t = 50 ms, t =
55 ms and t = 85 ms, respectively. While for the higher concentration
systems (φ = 2^.10^-4
and φ
= 5^.10^-4)
the top point of droplet approached to the equilibrium interface smoothly and
slowly, but at much longer times than at the lower concentrations. It is worth
mentioning that the top point in these cases was not reaching heights below the
final interface level as was the case with the lower surfactant systems.

Fig.
2. The
variation of the relative height h_s of the top point of the
droplet

The
velocity fields inside the droplet for different surfactant concentrations and
different time steps are presented in Figs. 3 and 4. Post interface rupture,
two counter-rotating vortices were observed at the bottom of the droplet. With
increasing time, the two counter-rotating vortices started to attach to the
upper part of the droplet while their intensities were gradually increased
until a transition point was reached, after which the intensity decreased. At
this initial stage the fluid dynamics inside the droplets were dominated by the
vortices present, while after the transition point the flow was dominated by the
downward axial fluid motion. Also, it was observed that for increasing surfactant concentration the intensity of the two
counter rotating vortices was decreased. The location of the two vortices
inside the droplet was also tracked and compared for different surfactant
concentrations and kinetic energy distributions for different surfactant
concentrations were computed.

Fig.
3.
Velocity fields and contour of the vorticity evolution for φ = 1^.10^-4

Fig. 4. Velocity fields and
contour of the vorticity evolution for φ = 2^.10^-4

Acknowledgements

The
project was funded by the UK Engineering and Physical Science Research Council
(EPSRC) Programme Grant MEMPHIS. T. Dong would also like to thank CSC and UCL
for his PhD studentship.