PAPER www.rsc.

org/pccp | Physical Chemistry Chemical Physics

Compositional ripening of particle- and surfactant-stabilised emulsions:

a comparison
Bernard P. Binks,a Paul D.I. Fletcher,*a Benjamin L. Holt,a Olivia Kuc,a
Pascal Beaussoubreb and Kenneth Wongb
Received 10th September 2009, Accepted 2nd December 2009
First published as an Advance Article on the web 18th January 2010
DOI: 10.1039/b918812f

When b-ionone-in-water emulsions are mixed with squalane-in-water emulsions, the slightly
water-soluble, mobile b-ionone undergoes mass transfer to the drops of highly water-insoluble,
immobile oil squalane. We have investigated this compositional ripening process for emulsions
stabilised either by particles or by surfactant molecules. For particle-stabilised emulsions, the
swelling of the squalane-containing drops triggers droplet coalescence which causes the final
swollen droplet radius to be proportional to the swelling ratio to the power of 1.
Surfactant-stabilised emulsions swell without coalescence which causes the final droplet radius
to be proportional to the swelling ratio to the power 1/3. Addition of excess, non-adsorbed
particles to the particle-stabilised emulsions suppresses the swelling-triggered coalescence and
causes a switchover from particle to surfactant behaviour.

Introduction second emulsion containing a mobile oil. In this case, the

Oil-in-water (o/w) emulsions are thermodynamically unstable
dispersions of micrometre-sized oil droplets in water which
may be kinetically stabilised by adsorbed films of either
surfactants or particles.1,2 When two stable o/w emulsions
containing different, but mutually miscible oils are mixed,
mass transfer of the oils can occur between the droplets to
produce a single population of mixed oil drops.3–25 The
process, denoted as compositional ripening, is driven by the
tendency of the two oils to mix and is related to Ostwald
ripening (OR) in which mass transfer is driven by Laplace
pressure differences between small and large droplets.26–29 For
both ripening processes, the mass transfer occurs by a
mechanism of dissolution of the oil from the droplet into
the aqueous continuous phase followed by diffusion and
re-dissolution back into an emulsion droplet. As noted in
ref. 21, the chemical potential difference arising from
concentration differences normally greatly outweighs that
due to Laplace pressure differences and so mass transfer under
the conditions used here is dominated by compositional
ripening and hence OR can be neglected. The rate of mass
transfer depends on the droplet surface area, the aqueous
phase solubility, the diffusion coefficient of the oil through
water and any possible energy barrier to transport across the
oil/water interfaces. In considering whether different oils will
transfer at significant rates, the dominant factor is the aqueous
phase solubility and so very water-insoluble oils such as
squalane show negligible mass transfer rates and can be
considered as immobile oils whereas moderately soluble oils
are mobile.21,22 In this paper we focus on systems in which an
o/w emulsion containing an immobile oil is mixed with a
a
Surfactant & Colloid Group, Department of Chemistry, University of
Hull, Hull, UK HU6 7RX. E-mail: P.D.Fletcher@hull.ac.uk
b
Firmenich SA, P.O. Box 239, Route des Jeunes 1,
CH-1211 Geneva 8, Switzerland
immobile oil drops swell and the mobile oil drops shrink as
mass transfer proceeds.
The main objective of the work described here is to compare
how this type of compositional ripening occurs in emulsions
stabilised solely by adsorbed particles with that in emulsions
stabilised by surfactants. Differences in the adsorption/desorption
behaviour of particles and surfactant are expected to lead to
differences in the compositional ripening behaviour of the two
types of emulsions as discussed below.
Theoretical considerations
For o/w emulsions stabilised by surfactants, the adsorbed
surfactant films stabilising the droplets are in dynamic
equilibrium with non-adsorbed monomeric surfactant (which
can be distributed between the aqueous and oil phases) and
with surfactant aggregates present in the aqueous phase
(normally micelles containing some solubilised oil or o/w
microemulsion droplets). The aqueous phase concentration
of non-adsorbed surfactant is generally approximately equal
to the critical micelle concentration (cmc) of the surfactant in
water in the absence of added oil.30 Because of the dynamic
equilibrium nature of the adsorption and the presence of a
reservoir of surfactant in the form of o/w microemulsion
aggregates, slow swelling or shrinkage of the emulsion drops
is accompanied by relatively rapid adsorption or desorption so
that the surface concentrations of the surfactant on the
swelling and shrinking emulsion drop surfaces are maintained
constant as their areas change. When this surface concentration
of surfactant is sufficient to prevent droplet coalescence, the
concentration of emulsion drops containing immobile oil is
maintained constant as they swell. Mass transfer of the mobile
oil will proceed to completion, i.e. all the mobile oil added
initially as a separate emulsion will be incorporated into the
final population of swollen drops containing a mixture of the

This journal is

c the Owner Societies 2010 Phys. Chem. Chem. Phys., 2010, 12, 2219–2226 | 2219
two oils. Hence, the radius of the emulsion drop population
containing immobile oil will swell from an initial value rinit to a
final value rfinal given by21
rfinal = rinit{(fimm + fmob)/fimm}1/3 (1)
where fimm is the volume fraction of immobile oil and fmob is
the volume fraction of mobile oil.
Unlike the case for surfactant adsorption, particle adsorption
at the oil/water surface is commonly expected to be irreversible.
For a spherical adsorbed particle which makes a contact angle y
with the oil/water interface, the energy of adsorption Eads from
either the oil or water phase is given by1
Eads = prp2g{1  cos y}2 (2)
where rp is the radius of the particle, g is the tension of the
oil/water interface and the + or sign signifies that adsorption
occurs from either the oil or water phase. Particle adsorption is
most favourable when the contact angle y = 901. For typical
values of g (in the range of 10–40 mN m 1) with rp larger than
about 10 nm and y fairly close to 901, the adsorption energy is
orders of magnitude larger than the particle thermal energy kT
(where k is the Boltzmann constant and T is the absolute
temperature) with the result that particle adsorption is
irreversible. With this in mind, we now consider the processes
that are likely to occur when we trigger compositional ripening
by mixing two populations of particle-stabilised o/w emulsions,
one containing drops of a mobile oil and the second containing
an immobile oil. We assume here that the concentration of
non-adsorbed particles is zero and that a certain critical
packing density of adsorbed particles around the emulsion
drops is required to stabilise them against coalescence. As
shown schematically in Fig. 1, mass transfer of the mobile oil
from the mobile drops into the drops of immobile oil serves to
compress the particle film around the mobile oil drops and
expand the particle films of the immobile oil drops. For the
immobile oil drops, the particle film expansion is expected to
lead to those drops becoming unstable with respect to coalescence
which then occurs leading to re-compression of the particle
films back to their original surface density such that they are
re-stabilised against coalescence. This swelling-triggered
coalescence is similar to the so-called limited coalescence of
related systems which has been reported.31 Hence, for the oil
drops containing immobile oil, the two processes of expansion
and coalescence combine to swell the radius of this drop
population from rinit to rfinal with conservation of the overall
surface area of this drop population. This behaviour leads to
the expectation that:
1
rfinal = rinit{(fimm + fmob)/fimm} (3)
The scaling power of 1 expected for irreversibly adsorbed
particle-stabilised emulsion drops for which the swelling drop
population surface area is conserved is in contrast to the power
of 1/3 expected for reversibly adsorbed surfactant-stabilised
systems in which the number of emulsion drops is conserved.
Eqn (3) is derived on the basis that all of the mobile oil
transfers from the initial mobile oil drop population into the
swollen/coalesced drop population. However, as the mobile oil
transfers, it shrinks the drops and compresses the adsorbed
2220 | Phys. Chem. Chem. Phys., 2010, 12, 2219–2226
Fig. 1 Schematic of the droplet swelling and coalescence processes
occurring during the compositional ripening of particle-stabilised
droplets. Note that, for clarity, the particles coating the droplets of
mobile oil are omitted here.
particle films from their initial surface density. A close-packed
film of solid particles will resist shrinkage which can only
be accommodated by either particle desorption (thought to
be energetically highly unfavourable, see eqn (2)) or by
‘‘crumpling’’ of the film.32,33 Hence, it may be the case that
the shrinking of the mobile oil drops does not proceed to
completion with the result that not all the mobile oil transfers.
Relevant to this point, it has been shown that films of
adsorbed particles coating gas bubbles under water can
successfully prevent bubble shrinkage due to gas dissolution
in the surrounding water phase.34,35
The compositional ripening behaviour described by eqn (3)
corresponds to the situation in which all the particles are
assumed to be adsorbed. However, if sufficient excess particles
are added, it is expected that the new oil/water interface
exposed by swelling of the emulsion drops may be filled by
adsorbing particles and hence no droplet coalescence will be
triggered. Under these conditions, the swelling behaviour
should switch from ‘‘particle-like’’ as described by eqn (3) to
‘‘surfactant-like’’ as described by eqn (1).
In this paper we describe results of experiments designed to
test these predictions. We observe the compositional ripening
which occurs when we mix o/w emulsions containing squalane
as the immobile oil with emulsions containing a moderately
polar mobile oil (b-ionone in most cases) with either particles
or a surfactant as stabiliser.
Experimental
Materials
The oils squalane (Aldrich, 99%), b-ionone (4-(2,6,6-
trimethyl-1-cyclohexenyl)-3-buten-2-one, Aldrich, 97+%) and
a-hexyl cinnamaldehyde (Aldrich, 95+%) were columned
three times over neutral alumina prior to use in order to
remove any polar impurities. Water was purified by reverse
osmosis and treated using a Milli-Q reagent water system.
Fumed silica nanoparticle samples contained primary particles
which were approximately spherical with diameters of 10–30 nm
but also contained fused aggregates and higher agglomerates
This journal is
c the Owner Societies 2010
with sizes ranging up to several hundred nm. The particles
were hydrophobised to different extents by treatment with
dichlorodimethylsilane and their hydrophobicity characterised
by measurement of the residual percentage of surface SiOH
groups. Thus, 100% SiOH corresponds to untreated, hydrophilic
silica whereas the most hydrophobic grade used here was 71%
SiOH. All silica particles were a generous gift from Wacker
Chemie, Burghausen, Germany. Nile red fluorescent dye
(Sigma-Aldrich, technical grade) was used to stain the
b-ionone emulsion drops for fluorescence microscopic imaging.
The non-ionic surfactant H–(CH2)12–(O–CH2–CH2)8–OH
(abbreviated to C12E8, Nikko Chemicals, Japan chromato-
graphically pure grade) was used as supplied.
Methods
Particle-stabilised emulsions were prepared by first dispersing
the silica particles in water using an ultrasonic probe
(Vibra-Cell, Sonics & Materials) using a 3 mm diameter probe
tip operating at 20 kHz with 10 W power for 2 min. The
sample was kept cool during sonication using an ice bath.
The volumes of oil and aqueous particle dispersion required to
produce the final, desired emulsion composition were
combined to give a total volume of 10 ml which was then
emulsified using an IKA Ultra-Turrax T25 homogeniser
fitted with an 18 mm diameter dispersing head operating at
13 000 rpm for 2 min. For the surfactant-stabilised emulsions,
the surfactant was dissolved initially in the water and suitable
volumes of this solution and the oil were combined to give a
total volume of 40 ml. The emulsions were homogenised using
a Silverson mixer (model L4R) operating at its lowest speed
for 1 min.
For the compositional ripening experiments with particle-
stabilised systems, the volumes of the two emulsions and water
(with or without additional dispersed particles) required to
give the required final overall composition were mixed to give
a total volume of 6 ml. The samples were contained in
stoppered glass sample tubes containing a magnetic flea and
stirred magnetically at the lowest stirring speed sufficient to
prevent visually observable sedimentation or creaming.
Surfactant-stabilised emulsions were incubated similarly
except that the mobile oil (b-ionone) was added directly as
the pure oil into the squalane-in-water emulsion.
To obtain optical micrographs of the emulsions, 2 drops of
the incubating sample were placed on a dimpled microscope
slide and covered with a cover slip. Microscope images were
obtained with an Olympus BX51 microscope operating in
either transmission or fluorescence mode and equipped with
an Olympus DP70 camera. Fluorescence images with nile red
as stain were obtained using an Olympus model UMF2 filter
set with excitation at 451–490 nm and emission at 491–540 nm.
The digital images were analysed using Image-Pro Plus
software and the scale calibrated using an NPL model 164
reference stage graticule. The size distributions of the two drop
populations in the mixed emulsion samples were obtained by
measurement of at least 50 drops for each sample. For both
types of drop population, all distributions were monomodal
and were parameterised using the mean drop diameter and the
standard deviation of the distribution.
This journal is
c the Owner Societies 2010
Scanning electron microscope (SEM) images were obtained
using a Zeiss EVO 60 instrument. A drop of the sample
emulsion placed on a clean glass slide was dried in air at
60 1C for 15 min. The dried samples were sputter coated
with carbon prior to SEM imaging.
All measurements were made at room temperature of
20  2 1C.
Results and discussion
For the first series of experiments, we observed the
compositional ripening on mixing a squalane-in-water
emulsion (immobile oil) stabilised by 71% SiOH silica
particles with a b-ionone-in-water emulsion (mobile oil)
stabilised by 87% SiOH silica particles. The particle
concentrations were chosen to be the minimum required to
produce emulsions which were stable with respect to
coalescence (determined by checking that there was no visible
phase separation of the drop phase from the emulsion after
incubation for 1 week) and so the emulsions are considered to
contain virtually no non-adsorbed particles. Fig. 2 shows a
selection of optical micrographs taken during the time
course of the compositional ripening for a swelling ratio
(fimm + fmob)/fimm = 2. The two micrographs for t = 1 minute
correspond to transmission and fluorescence micrographs of
the same sample and show the identities of the two droplet
populations as the unstained squalane drops show as black
and the stained b-ionone drops show as red in fluorescence. As
time progresses, the squalane drops swell and the b-ionone
drops shrink progressively and the two droplet types are
readily identified by their different sizes using transmission
micrographs alone. After about 80 minutes, the compositional
ripening is complete and the two t = 90 minute micrographs
show two transmission images of the same sample at different
focal planes (and magnifications). The swollen drop popula-
tion contains a significant fraction of slightly non-spherical
drops (t = 90 min, left hand side image) and the origin of this
effect will be discussed later. After 90 minutes the b-ionone
drops have lost virtually all their oil and remain only as
deformed, aggregates consisting mainly of silica particles
(t = 90 min, right hand side image). Evidence for this point
arises not only from the small sizes of the residue particles
but also from visual observations of their sedimentation
behaviour in the mixed emulsions. Initially, the silica-stabilised
b-ionone drops contain a high fraction of b-ionone
(density = 0.945 g cm 3) and a low fraction of silica
(density = 2.2 g cm 3) and so the silica-coated drops are
initially less dense than the water continuous phase. Hence they
cream to the top of the emulsion unless stirred. As compositional
ripening proceeds, the fraction of silica increases and the
silica-coated droplet density increases until a point is reached
when the droplets sediment rather than cream. The residue
droplets seen in the t = 90 minutes micrograph are observed
to sediment within a few seconds in the microscope cell.
Fig. 3 (upper plot) shows the variation of the mean drop size
for both the swelling and the shrinking drop populations for
the experimental run of Fig. 2. It can be seen that both drop
sizes reach constant values after about 80 minutes or so. As
shown in Fig. 3 (lower plot), it was checked that incubation
Phys. Chem. Chem. Phys., 2010, 12, 2219–2226 | 2221
Fig. 2 Optical micrographs showing the time course of the compositional ripening of a mixed o/w emulsion containing fimm (squalane) of 4.2%
stabilised by 0.083 wt% of 71% SiOH and fmob (b-ionone, stained with nile red) of 4.2% stabilised by 0.083 wt% of 87% SiOH. All images are
transmission micrographs except for the red t = 1 min which is a fluorescence image. All scale bars represent 100 mm and the t values refer to the
time after mixing of the two emulsions.

with stirring (but with no compositional ripening) of the
squalane-in-water emulsion produced no significant droplet
size change over the time of the experiment.
Fig. 4 shows the time course of the compositional ripening
swelling of squalane-in-water drops stabilised by the non-ionic
surfactant C12E8 with the same swelling ratio (fimm + fmob)/
fimm = 2 as used in Fig. 3. The mobile oil b-ionone was added
by stirring in the pure oil (not as a separately prepared
emulsion) and so shrinking drop sizes were not measured.
Again, a control measurement of the drop sizes with incubation
time showed no change in the absence of compositional
ripening. It can be seen that the final extent of swelling in this
surfactant-stabilised emulsion is less than for the particle-
stabilised system of Fig. 3. In contrast to the swollen

2222 | Phys. Chem. Chem. Phys., 2010, 12, 2219–2226 This journal is

c the Owner Societies 2010

Fig. 3 The upper graph shows the variation of the radii of the
swelling and shrinking droplets for the system of Fig. 2. The lower
plot shows the variation of squalane droplet radius in a control
emulsion with no added droplets of b-ionone. The vertical bars
indicate the standard deviations of the droplet size distributions.
particle-stabilised emulsions, optical micrographs of the final
swollen surfactant-stabilised emulsions (not shown) showed
no non-spherical droplets.
Fig. 4 The variation of the radius of the swelling droplets for a mixed
emulsion containing fimm = 4.2% (squalane) and fmob = 4.2%
(b-ionone) with 0.5 wt% of the stabilising surfactant C12E8. The
vertical bars indicate the standard deviations of the droplet size
distributions.
This journal is
c the Owner Societies 2010
Fig. 5 summarises the compositional ripening swelling
behaviour of various particle-stabilised systems with squalane
as immobile oil and either b-ionone or a-hexyl cinnamm-
aldehyde as mobile oil (filled points). The swelling ratio
(fimm + fmob)/fimm was varied over the range 1–10 by
changing the ratio of volume fractions of the two oils at fixed
total oil volume fraction and by changing the total oil volume
fraction. For swelling ratios of less than approximately 3
(indicated by the vertical dashed line in Fig. 5), the predicted
behaviour of eqn (3) is followed. For larger swelling ratios, it
was found that the large drops in the squalane-containing
population were mostly unstable with respect to coalescence
and produced some macroscopic oil phase separation leaving
some smaller residual squalane-containing drops whose sizes
are plotted in Fig. 5. This coalescence instability of large,
swollen drops is consistent with the observation that emulsion
drops generally become more unstable with respect to
coalescence for increasing drop sizes.2 Fig. 5 also shows the
comparison with results for a range of surfactant-stabilised
systems which were mainly taken from ref. 23 but also include
the data for the squalane–b-ionone–C12E8 system measured
here. The surfactant systems follow the prediction of eqn (1)
with reasonable precision and are clearly distinguishable from
the particle-stabilised systems.
Shrinkage of the b-ionone drops in the particle-stabilised
emulsion systems produces deformed aggregates containing
little or no residual b-ionone as seen in the t = 90 minutes
optical micrographs of Fig. 2. Fig. 6 shows higher magnification
SEM images of the particle aggregates obtained by shrinking
b-ionone drops stabilised by 100% SiOH silica particles and
with an initial mean drop diameter of 20.6 mm. Images of
broken particle fragments (see Fig. 6(c)) indicate that they
consist of rough, thickened particle shells of thickness of
Fig. 5 Variation of final droplet radius with swelling ratio for
particle-stabilised emulsions (filled points) and surfactant-stabilised
emulsions (open points). Diamonds refer to squalane–b-ionone
with either 87% SiOH particles or C12E8, squares refer to
squalane–b-ionone with either 100% SiOH particles or surfactant
(data from ref. 23) and the filled triangle refers to squalane–a-hexyl
cinnamaldehyde with 78% SiOH particles. The dashed line
corresponds to the prediction of eqn (1) and the solid line to
eqn (3). The vertical error bars correspond to an estimated uncertainty
in the mean radius ratio (rfinal/rinit) of 10%.
Phys. Chem. Chem. Phys., 2010, 12, 2219–2226 | 2223
Fig. 6 SEM images of 100% SiOH silica-stabilised b-ionone emulsion drops after drying on a glass slide. Images (a)–(c) show their appearance
after a mean volume loss of 80% of the internal oil after mixing with an equal volume of squalane emulsion drops. Image (d) shows the original
emulsion drops without any shrinkage which have collapsed during drying. Scale bars represent 10 mm.

the order of 1 mm, considerably larger than the thickness It might be argued that the distinction drawn here between
(approx. 100 nm) of the particle shells (consisting of particle and surfactant behaviour might be better formulated
monodisperse, partially-hydrophobised silica particles of 25 in terms of systems with or without excess non-adsorbed
nm diameter) of intact, non-shrunken, frozen emulsion drops
imaged using a cryo-SEM method.36 A further noteworthy
point is that these objects are sufficiently stable to mostly
remain as 3-D entities following the sample preparation for
SEM. When a sample of the original, non-shrunken emulsion
was prepared and imaged by SEM, the droplet particle shell
structure collapsed onto the support to give the SEM image
shown in Fig. 6(d).
We next examined the effect of adding excess particles on
the compositional ripening behaviour of particle-stabilised
emulsions. Fig. 7 compares the time course of the mean radius
of the swelling drops in the absence and presence of excess
particles for a swelling ratio (fimm + fmob)/fimm = 2. In the
absence of excess particles, the drop diameter swells by
approximately 3-fold, slightly larger than that predicted by
eqn (3). In the presence of excess particles, the droplets swell
by a factor of approximately 1.4 which is reasonably close to
Fig. 7 Variation of swelling drop radius with time for emulsions
the factor 1.26 expected from eqn (1). Semi-quantitatively at
initially containing 4.2 vol% of b-ionone drops stabilised by
least, this experiment confirms the expectation that the 0.083 wt% of 87% SiOH particles and 4.2 vol% of squalane drops
addition of excess particles, expected to adsorb on the bare stabilised by 0.083 wt% of 71% SiOH particles with the addition of
oil/water interface created by droplet swelling and so prevent either 0 (filled diamonds), 0.209 wt% (open diamonds or 0.834 wt%
coalescence, causes a switchover from particle to surfactant (filled triangles) of 71% SiOH particles. The vertical bars indicate the
behaviour. standard deviations of the droplet size distributions.

2224 | Phys. Chem. Chem. Phys., 2010, 12, 2219–2226 This journal is

c the Owner Societies 2010
Fig. 8 Transmission optical micrographs of the final emulsion obtained after complete compositional ripening of the system of Fig. 7
containing excess, non-adsorbed particles (0.209 wt%). Scale bars represent 100 mm. Image (a) shows the swollen drops and (b) shows the
shrunken drops.
stabilising materials, either particles or surfactants. However,
common low molar mass surfactants commonly require mM
concentrations (similar in magnitude to the critical micelle
concentration in aqueous systems) of non-adsorbed,
monomers in the water, oil or both phases to produce
sufficient adsorption at the oil/water interface to stabilise
emulsions. For particles with adsorption energies which are
normally orders of magnitude larger than the thermal energy
kT, the concentration of non-adsorbed particles in equilibrium
with a saturated monolayer is expected to be vanishingly
small. Hence, common surfactant systems normally contain
significant concentrations of non-adsorbed species whereas
particle systems do not. It remains an open question whether
‘‘particle-like’’ behaviour might be found for systems containing
phospholipid or polymeric surfactants for which the
equilibrium concentrations of non-adsorbed species (similar
to their critical micelle concentration) are vanishingly small.
We return to the observation that swelling of particle-
stabilised droplets leads to the formation of non-spherical
droplets (t = 90 min image of Fig. 2). The origin of this effect
can be realised by considering the hypothetical coalescence of
2 droplets as triggered by an infinitesimally small expansion of
their surface areas. In this case, the surface area of adsorbed
particle film (virtually equal to that of the 2 original drops) will
be too large for a single, coalesced drop which is spherical.
The excess area is then accommodated by the adoption of a
non-spherical shape. A similar effect has been observed for
particle-stabilised bubbles.37 In principle, analysis of the axial
ratios of the swollen and coalesced droplets on completion of
compositional ripening could be used to extract information
about the extent of film expansion required to trigger coalescence.
However, such an analysis is (greatly) complicated by the
polydispersity of the drops and is not attempted here.
Instead, we focus on a comparison of the drop shapes of the
particle-stabilised emulsions which have been compositionally-
ripened in the absence and presence of excess particles.
Comparison of the micrographs of Fig. 2 and 8 shows that non-
spherical drops are not produced in the presence of excess particles
since these suppress the coalescence following droplet swelling.
This journal is
c the Owner Societies 2010
Finally, it is noteworthy that the results described here
show that droplet shrinking and expansion induced by
compositional coarsening proceeds to the final drop sizes
successfully explained using the simple model presented here.
In principle, one might envisage that the dilational properties
of the adsorbed particle films might suppress the mass transfer,
as has been shown for the case of mass transfer induced by
Laplace pressure differences between particle- or protein-
stabilised gas bubble and emulsion drop systems.38–42 A likely
cause of this difference in behaviour is that the driving force
for mass transfer is much larger in the case of compositional
ripening than for Laplace pressure differences.21 It is also
likely that dilational properties of the adsorbed particle films
affect the rate of the mass transfer leading to drop shrinkage
and swelling. In the present system, the mass transfer rates are
convoluted with drop coalescence and so further analysis,
although interesting, has not been attempted here.
Conclusions
The main conclusions from this work are:
K Compositional ripening of particle-stabilised emulsions
in the absence of excess, non-adsorbed particles causes
swelling which triggers droplet coalescence. The net effect of
these processes is that the final swollen droplet diameter scales
with swelling ratio to the power 1. This scaling relationship is
followed at low swelling ratios but breaks down for large
swelling ratios.
K Compositional ripening of surfactant-stabilised
emulsions causes swelling in which the final droplet diameter
scales with swelling ratio to the power 1/3.
K For the drops which shrink during compositional
ripening of particle-stabilised emulsions, it appears that most
or all the mobile oil is extracted. The particles which remain
consist of deformed, thickened shells.
K Addition of excess, non-adsorbed particles during
compositional ripening of particle-stabilised emulsions causes
the behaviour to switch from particle-like to surfactant-like.
Phys. Chem. Chem. Phys., 2010, 12, 2219–2226 | 2225
 K Particle-stabilised droplet coalescence triggered by droplet
swelling leads to the formation of non-spherical droplets.
Addition of excess particles during swelling suppresses drop
coalescence and so retains spherical droplets.
Acknowledgements
We thank Firmenich SA for a fully funded studentship award
to BLH.
Notes and references
1 B. P. Binks, Curr. Opin. Colloid Interface Sci., 2002, 7, 21.
2 S. Tcholakova, N. D. Denkov and A. Lips, Phys. Chem. Chem.
Phys., 2008, 10, 1608.
3 G. W. Hallworth and J. E. Carless, J. Pharm. Pharmacol., 1972,
24(suppl.), 71P.
4 R. Buscall, S. S. Davis and D. C. Potts, Colloid Polym. Sci., 1979,
257, 636.
5 S. S. Davis, H. P. Round and T. S. Purewal, J. Colloid Interface
Sci., 1981, 80, 508.
6 A. V. Pertsov, A. S. Kabalnov and E. D. Shchukin, Kolloidn. Zh.,
1984, 46, 1172.
7 A. S. Kabalnov, A. V. Pertzov, Y. D. Aprosin and E. D. Shchukin,
Colloid J. USSR, 1985, 47, 898.
8 A. S. Kabalnov, E. D. Shchukin and A. V. Pertsov, Colloids Surf.,
1987, 24, 19.
9 A. V. Pertsov, A. S. Kabalnov, E. E. Kumecheva and
E. A. Amelina, Kolloidn. Zh., 1988, 50, 543.
10 E. E. Kumecheva, E. A. Amelina, A. V. Pertsov and
E. D. Shchukin, Kolloidn. Zh., 1989, 51, 1059.
11 E. E. Kumecheva, E. A. Amelina and A. M. Parfenova, Kolloidn.
Zh., 1990, 52, 318.
12 A. S. Kabalnov and E. D. Shchukin, Adv. Colloid Interface Sci.,
1992, 38, 69.
13 D. J. McClements and S. R. Dungan, J. Phys. Chem., 1993, 97,
7304.
14 D. J. McClements, S. R. Dungan, J. B. German and J. E. Kinsella,
J. Colloid Interface Sci., 1993, 156, 425.
15 D. J. McClements, S. R. Dungan, J. B. German and J. E. Kinsella,
Colloids Surf., A, 1993, 81, 203.
16 J. G. Weers and R. A. Arlauskas, Langmuir, 1995, 11, 474.
17 R. A. Arlauskas and J. G. Weers, Langmuir, 1996, 12, 1923.
18 L. Taisne, P. Walstra and B. Cabane, J. Colloid Interface Sci.,
1996, 184, 378.
2226 | Phys. Chem. Chem. Phys., 2010, 12, 2219–2226
19 J. Weiss, J. N. Coupland and D. J. McClements, J. Phys. Chem.,
1996, 100, 1066.
20 A. Kabalnov and J. Weers, Langmuir, 1996, 12, 3442.
21 B. P. Binks, J. H. Clint, P. D. I. Fletcher, S. Rippon,
S. D. Lubetkin and P. J. Mulqueen, Langmuir, 1998, 14, 5402.
22 B. P. Binks, J. H. Clint, P. D. I. Fletcher, S. Rippon,
S. D. Lubetkin and P. J. Mulqueen, Langmuir, 1999, 15, 4495.
23 P. J. Mulqueen, S. D. Lubetkin, G. W. Smith and E. S. Paterson,
United States Patent, 6 017 559, 2000.
24 A. A. Pena and C. A. Miller, J. Colloid Interface Sci., 2001, 244,
154.
25 A. A. Pena and C. A. Miller, Adv. Colloid Interface Sci., 2006,
123–126, 241.
26 P. Taylor, Colloids Surf., A, 1995, 99, 175.
27 J. Weiss, C. Canceliere and D. J. McClements, Langmuir, 2000, 16,
6833.
28 A. Kabalnov, J. Dispersion Sci. Technol., 2001, 22, 1.
29 P. Taylor, Adv. Colloid Interface Sci., 2003, 106, 261.
30 R. Aveyard, B. P. Binks, S. Clark and P. D. I. Fletcher, J. Chem.
Soc., Faraday Trans., 1990, 86, 3111.
31 S. Arditty, C. P. Whitby, B. P. Binks, V. Schmitt and F.
Leal-Calderon, Eur. Phys. J. E, 2003, 11, 273; S. Arditty,
C. P. Whitby, B. P. Binks, V. Schmitt and F. Leal-Calderon,
Eur. Phys. J. E, 2003, 12, 355.
32 T. S. Horozov, B. P. Binks, R. Aveyard and J. H. Clint, Colloids
Surf., A, 2006, 282–283, 377.
33 R. Aveyard, J. H. Clint, D. Nees and V. N. Paunov, Langmuir,
2000, 16, 1969.
34 Z. P. Du, M. P. Bilbao-Montoya, B. P. Binks, E. Dickinson,
R. Ettelaie and B. S. Murray, Langmuir, 2003, 19, 3106.
35 T. Kostakis, R. Ettelaie and B. S. Murray, Langmuir, 2006, 22,
1273.
36 B. P. Binks, J. H. Clint, A. K. F. Dyab, P. D. I. Fletcher,
M. Kirkland and C. P. Whitby, Langmuir, 2003, 19, 8888.
37 A. B. Subramaniam, M. AbKarian, L. Mahadevan and
H. A. Stone, Langmuir, 2006, 22, 10204; A. B. Subramaniam,
M. AbKarian, L. Mahadevan and H. A. Stone, Nature, 2005, 438,
930.
38 M. A. Bos and T. van Vliet, Adv. Colloid Interface Sci., 2001, 91,
437.
39 M. B. J. Meinders, W. Kloek and T. van Vliet, Langmuir, 2001, 17,
3923.
40 E. Dickinson, R. Ettelaie, T. Kostakis and B. S. Murray,
Langmuir, 2004, 20, 8517.
41 M. Abkarian, A. B. Subramaniam, S. H. Kim, R. J. Larsen,
S. M. Yang and H. A. Stone, Phys. Rev. Lett., 2007, 99, 188301.
42 A. Stocco, W. Drenckham, E. Rio, D. Langevin and B. P. Binks,
Soft Matter, 2009, 5, 2215.
This journal is
c the Owner Societies 2010