Current Opinion in Colloid & Interface Science 15 (2010) 40–49

Contents lists available at ScienceDirect

Current Opinion in Colloid & Interface Science

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c o c i s

Food emulsions and foams: Stabilization by particles

Eric Dickinson ⁎
School of Food Science and Nutrition, University of Leeds, Leeds LS2 9JT, UK

a r t i c l e i n f o a b s t r a c t

Article history: Recent advances in the stabilization of emulsions and foams by particles of nanoscale and microscopic
Received 28 August 2009 dimensions are described. Ongoing research in this highly active field is providing insight into (i) the
Accepted 9 November 2009 molecular factors controlling particle wettability and adsorption, (ii) the structural and mechanical
Available online 14 November 2009
properties of particle-laden liquid interfaces, and (ii) the stabilization mechanisms of particle-coated
droplets and bubbles. There is much potential for exploiting the emerging knowledge in new food product
Keywords:
Emulsions
applications. The preparation of cheap and effective colloidal particles based on food-grade ingredients,
Foams especially proteins, is the key technological challenge.
Droplets © 2009 Elsevier Ltd. All rights reserved.
Bubbles
Nanoparticles
Proteins
Pickering stabilization
Contact angle
Wettability
Silica
Fat crystals
Surfactants
Emulsifiers

1. Introduction previous reviews [2–12•]. Attention here is directed towards research

published during the past 3–4 years. The purpose of this article is
Food systems commonly contain particulate material that accu- focus on recent developments in (nano)particle stabilization that
mulates at oil–water and air–water interfaces and contributes to the seem particularly relevant to food emulsions and foams. The
colloidal stabilization of emulsions and foams [1•]. The range of perceived relevance may be because the investigators have worked
particle sizes involved in structuring such systems during and after directly with food-grade ingredients, or, alternatively, because the
emulsification and foaming can vary widely from just a few research offers potential strategies or methodologies for developing
nanometres to tens of micrometres. As applied to emulsion droplets improved (nano)particle-based systems in food product formulations.
coated by a layer of adsorbed solid particles at the oil–water interface, This article is mainly concerned with experimental investigations of
the mechanism is commonly referred to as Pickering stabilization. Oil- dispersed systems stabilized by inorganic particles (especially silica
in-water (O/W) or water-in-oil (W/O) emulsions can be produced nanoparticles), fat crystals, and protein-based nanoparticles. Struc-
depending on whether the particles are predominantly hydrophilic or tures primarily based on supramolecular networks/gels or self-
hydrophobic. Important examples of Pickering-type food emulsions assembled lipids/surfactants are considered to lie outside the scope
are homogenized and reconstituted milks (O/W emulsions stabilized of the review.
by casein micelles) and margarines and fatty spreads (W/O emulsions
stabilized by triglyceride crystals). Examples of edible foams stabi- 2. Emulsions
lized by particles include whipped dairy cream and dessert toppings
stabilized by partially aggregated emulsion droplets [1•]. According to the conventional explanation for emulsion stabiliza-
During the past decade there has been increasing interest by tion by solid particles [1–3], there is an accumulation of particles at
physical scientists in the interfacial properties of particles, especially the oil–water interface in the form of a densely packed layer. This
nanoparticles. Some of this progress has been well described in thick particle layer then prevents droplet flocculation and coalescence
by a steric mechanism.
When two particle-coated droplets come close together we can
⁎ Tel.: + 44 113 233 2956; fax: + 44 113 233 2982. envisage a sort of particle bilayer arrangement as shown schemati-
E-mail address: E.Dickinson@leeds.ac.uk. cally in Fig. 1(a). The effectiveness of the physical barrier in inhibiting

1359-0294/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.cocis.2009.11.001
 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49 41

hindrance effect, which prevents significant particle displacement

away from or within the bridging layer.
Most particles have fairly uniform surfaces (i.e., they are not
amphiphilic). This implies that the surfaces of particle-coated droplets
have properties similar to those of the particles themselves. For
contact angles favouring strong adsorption at the oil–water interface,
this typically means that particles are in a state of weak aggregation.
Therefore a third type of interdroplet structural arrangement is
commonly encountered in which the droplets are stabilized through
the adsorption of aggregated colloidal particles [14,16•]. In this case, as
illustrated in Fig. 1(c), the steric particle-based barrier is not a simple
bilayer (or monolayer); nor is it necessarily densely packed. Rather it
consists of a rigid disordered layer/network of particles, adsorbed to
the oil–water interface(s), with the whole aggregated structure held
together by attractive interparticle forces.
The idealized model involving the stabilization of a Pickering
emulsion by a monolayer of uniform spherical particles is rarely
realized in practice. Sometimes this is simply because the particles are
highly polydisperse. However, even when the primary solid particles
are reasonably monodisperse, full dispersion of the monomer
particles in the liquid medium is apparently difficult to achieve
experimentally, even following high-pressure homogenization or
high-intensity ultrasound treatment. This is especially the case for
nano-sized particles, where the large surface area (high surface
energy) makes complete dispersion from the powder state an
unfavourable process thermodynamically. There are numerous
reports [3,4,9,11,14,16•] of emulsification studies involving silica
particles of nominal primary particle size 10–20 nm. But the actual
particulate entities involved in stabilizing the Pickering emulsions are
not the primary nanoparticles themselves: they are silica aggregates
of mean diameter 100–200 nm (based on laser light scattering) and
Fig. 1. Schematic representation of alternative stabilizing arrangements of particles in a having irregular sizes and shapes (based on electron microscopy). For
thin film between closely approaching emulsion droplets: (a) separate closely packed such polydisperse non-spherical aggregates, the proposition that
monolayers on the two oil–water interfaces; (b) single dense layer of bridging particles; there is just a single thermodynamic parameter (the three-phase
and (c) low-density network layer of aggregated particles.
contact angle at the particle surface) determining the character of the
interfacial layer structure seems difficult to accept.
coalescence is enhanced by the stability of the thin film of continuous Food emulsions are compositionally complex. Their droplets are
phase between the particle-coated droplets, as determined by the stabilized to differing extents by proteins, small-molecule surfactants
capillary pressure in the film and the surface rheological properties (emulsifiers), and, in certain cases, polysaccharides (hydrocolloids).
[4]. The magnitude of the steric barrier depends on how difficult it is In terms of the underlying mechanisms of (de)stabilization, there are
to remove particles from the interface. The steric barrier is more some similarities and differences between (nano)particles and the
effective when the particle layer lies predominantly on the outer other types of emulsifying agents [12,17,18]. When the concentration
(convex) side of the dispersed droplets, i.e., when the particles are of emulsifying agent is low, and electrostatic interactions are largely
preferentially wetted by the continuous phase. Particle location at the suppressed, the initially formed droplets tend to coalesce fairly
interface depends on the contact angle θ. For spherical particles of rapidly, until their surfaces are protected by a stabilizing layer of
radius r, the free energy of spontaneous desorption (ΔGD) is particles/molecules. This is usually accompanied by some bridging
proportional to r2(1–cosθ)2. This means that, so long as θ is not too flocculation due to sharing of adsorbed particles/macromolecules
close to 0° (or 180°), the predicted value of ΔGD for a particle of amongst neighbouring droplets. With a high concentration of
colloidal size is extremely large compared with the thermal energy emulsifying agent under controlled hydrodynamic conditions of
(kT); hence the particle is effectively permanently adsorbed. Even a emulsification, the resulting droplet size is mainly determined by
relatively small nanoparticle (r ∼ 5–10 nm), can be regarded as being the interfacial tension. The relevant value of the tension here is the
essentially irreversibly adsorbed, so long as the contact angle is not equilibrium value for rapidly adsorbing surfactants, but it is the
too far from 90°. dynamic tension for slowly adsorbing biopolymers (proteins and
Experimentally it has been observed [13–16•] that particle- hydrocolloids). For (nano)particle emulsifiers (acting alone), the
stabilized emulsion systems can be produced without a full mono- relevant interfacial tension is the (high) value at the bare oil–water
layer coverage of particles around the droplets. Therefore, in addition interface [12•]. This is the primary reason why it is difficult to make
to the classical arrangement of two densely packed monolayers particle-stabilized emulsions of small droplet size. Conventional
separating adjacent droplets (Fig. 1(a)), we need to consider other (macro)emulsions are recognized as being metastable, in contrast to
possible particle-based structures with the capacity to prevent droplet microemulsions which are thermodynamically stable (though the
coalescence. These structures typically involve some aggregation of formation of thermodynamically stable Pickering emulsions has in
particles or flocculation of the droplets. For instance, there could be fact been reported [19•]). Whatever is the character of the emulsifying
just a single dense layer of bridging particles (see Fig. 1(b)), where agent, however, once the emulsion is made, the main factor
each particle is wetted by both dispersed phase regions, but still with determining its coalescence stability is the nature of the repulsive
the major portion of each particle lying in the aqueous continuous interactions between the droplet surfaces. Except at low ionic
phase [4,15]. As with the particle bilayer arrangement, coalescence strength, the primary influence on the range of the interdroplet
stability of the single-layer configuration in Fig. 1(b) is due to a steric repulsive interactions is the thickness of the stabilizing layer, as
42 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49

determined by the relative physical size of the adsorbed species was apparently limited by the emulsification process. Under optimized
(particles N hydrocolloids N proteins N surfactants) [18]. conditions, the mixed emulsifier system was found to produce rather
To produce effective stabilization of O/W emulsions, it has been coarse emulsions, which were highly unstable to creaming, but
found that the surface character of hydrophilic silica particles has to be possessed long-term stability against coalescence. A two-stage syner-
made somewhat hydrophobic. For the purpose of controlling rheolog- gistic mechanism was proposed [28•] to explain the coalescence
ical properties, the required surface hydrophobic modification could be stability. It was assumed that the function of the small-molecule
achieved by lowering the pH of the aqueous medium (to pH∼ 2) for surfactant was to delay coalescence and to induce droplet break-up
nano-sized particles [20] or by adding multivalent cations (La3+) to the during emulsification, via its rapid adsorption and lowering of the
medium for larger particles (∼200 nm) [21]. Whether the droplet interfacial tension. This short-term interfacial stabilization by the
surface is coated by a dense layer of particles or is only sparsely covered monoolein allows time for the silica particles to assemble at the oil–
can be sensitively influenced by adjusting the pH and/or ionic strength water interface and thereby provide the long-term storage stability.
[16•]. The strategy of modifying the surface hydrophobicity by adjusting One way of avoiding having to add any small-molecule surfactant
the surface charge density was also implemented [22] by binding to modify the particle wettability is to stabilize an emulsion by
potassium hydrogen phthalate to positively charged alumina-coated heteroaggregation of a binary mixture of oppositely charged nano-
silica nanoparticles at 3.5 b pH b 5.5. particles. Binks et al. [29•] used this technique to make coarse gel-like
An obvious way to modify the surface hydrophobicity of colloidal O/W emulsions. They found that only those particle mixtures giving
particles is through adsorption of low-molecular-weight charged rise to flocculated aqueous dispersions were of low enough net charge
surfactants, e.g., anionic sodium dodecyl sulfate (SDS) or cationic and sufficiently hydrophobic to adsorb around oil droplets and confer
hexadecyltrimethylammonium bromide (CTAB). The balance of stability. Another strategy is to use solid particles that are themselves
interactions in the resulting emulsion system is, however, then amphiphilic — so-called ‘Janus particles’ (after the two-faced Roman
potentially more complicated because the surfactant can adsorb at god of doorways). These non-uniform particles are characterized by
both the liquid interface and the solid particle surface. Hence the two different hemispherical or surface regions, one polar and the
attachment strength of the particle to the interface depends on the other non-polar, with two different contact angles. Janus particles
extent to which the surfactant affects both the interfacial tension and have considerable advantages for use in emulsion stabilization and
the contact angle. For an O/W emulsion (50 vol.% n-dodecane) nanotechnology, but they are more difficult to make than conven-
containing mixtures of nanoparticles and oppositely charged surfac- tional (nano)particles [5].
tants, Binks et al. [23,24•] showed that the most stable coarse droplets The effect of particle shape on the coalescence stability of Pickering
were produced under conditions where the amount of adsorbed emulsions has been investigated recently by Madivala et al. [30•]. For
surfactant was sufficient to optimize the particle wettability and both O/W and W/O emulsions, the authors observed a strong
maximize the degree of particle flocculation. For systems based on dependence of emulsion stability on the particle aspect ratio in the
negatively charged silica particles + cationic CTAB [23] or positively range from 1 to 9. It appears that sufficiently long particles are effective
charged alumina-coated silica particles + anionic SDS [24•], the most emulsifying agents when spherical or less elongated particles of
stable emulsions (with respect to creaming and coalescence) were similar wettability cannot produce a stable emulsion. The behaviour
found to possess a gel-like consistency, with microstructural stability was rationalized [30•] in terms of the effect of particle aspect ratio on
enhanced by the presence of particle flocs adsorbed at the oil–water surface coverage and shape-induced attractive capillary interactions,
interface. Whether the oil–water interfacial layer has a liquid-like or as reflected in measurements of the interfacial shear rheology at the
solid-like character can be sensitively controlled by adjusting the planar oil–water interface.
surfactant/particle ratio, as demonstrated by optical scanning tomog- Oil-continuous emulsions can be stabilized by wax particles, both
raphy on W/O emulsions containing CTAB + silica nanoparticles [25]. natural (e.g., beeswax) and synthetic (polyethylene-based). Rousseau
In another study involving addition of SDS to Pickering O/W and Hodge have reported [31,32] that paraffin wax crystals can
emulsions after emulsification, it was observed [26] that low enhance the coalescence stability of W/O emulsions made with an oil-
concentrations of surfactant had no measurable effect on emulsion soluble surfactant as the primary emulsifying agent. The stabilization
stability, but at concentrations above the critical micelle concentra- mechanism involves wax particle adsorption and the formation of a
tion there was rapid creaming and flocculation. particle network in the continuous oil phase which immobilizes
Lecithin, a zwitterionic surfactant, is one of the most important coarse water droplets thereby preventing sedimentation. Emulsion
emulsifiers routinely used in food applications. The long-term properties were found to be dependent on whether the wax crystals
stability of fine O/W emulsions (10 vol.% triglyceride oil) prepared were present in the oil phase prior to emulsification at ambient
with silica nanoparticles + soybean lecithin (or oleylamine) was temperature, or whether they were formed in situ by rapid cooling
investigated by Eskandar et al. [27•]. These authors reported that a following high-temperature emulsification [31,32]. Using the latter
synergistic improvement in coalescence stability was evident only procedure, the crystals were smaller and the resulting emulsions were
when the nanoparticles were initially added to the oil phase. Under more stable. In a separate study, Binks and Rocher [33•] investigated
such circumstances, the free energy of particle adsorption estimated the application of various types of commercial wax particles in the
from measurements of interfacial tension and contact angles could be stabilization of W/O emulsions (50 vol.% squalene). While these
successfully correlated with the long-term stability. formulations contained no added surfactant, it would appear that the
Also of particular relevance to food applications are Pickering adsorption of surface-active molecules from the melted wax made a
emulsion systems containing non-ionic surfactants. The creaming and substantial contribution to emulsion stability, over and above the
coalescence stability of O/W emulsions (20 vol.% vegetable oil) prepared stability contributed by the wax particles themselves.
with a mixture of hydrophilic silica particles (primary particle size Dispersed fat crystals are the essential structural components of
∼12 nm, aggregate size ∼150 nm) and food-grade emulsifier mono- edible fatty spreads [1•]. After crystallization from the partially
olein (HLB= 3.8) was recently investigated as a function of pH and saturated oil phase, the triglyceride crystals interact and aggregate
system composition [28•]. In the absence of silica particles, the to form a three-dimensional network which provides long-term
emulsions prepared with monoolein as sole emulsifier were highly stability to the dispersed water droplets and a solid-like textural
unstable, exhibiting complete phase separation at pH= 2 after a couple character to the food product [34]. Triglyceride crystals adsorbing at
of hours. With the mixed emulsifier system, the concentrations of both the oil–water interface contribute to W/O emulsion stability by the
monoolein and silica particles were found to affect the emulsion droplet Pickering mechanism. The association of the fat crystals with the
size, up to some threshold concentrations beyond which the droplet size interface increases the surface viscoelasticity, especially when the
 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49 43

adsorbed crystals are flocculated. Addition of small-molecule emulsi- bicontinuous morphology is abruptly stabilized by jamming once a
fiers influences the fat crystal morphology, the strength of the crystal– certain surface concentration of particles has been reached. Analogous
crystal interactions, and the wetting properties of the crystals at the to the interfacial jamming in conventional O/W and W/O emulsion
oil–water interface [1•]. When the interparticle interactions are net systems, evidence for the same phenomenon has also been recently
attractive, and the particle concentration is sufficient to form a gel-like found [53•] in a bicontinuous water-in-water (W/W) emulsion based on
network with a finite yield stress, the destabilizing processes of a phase-separating aqueous solution of gelatin + starch. In this mixture
sedimentation and flocculation are inhibited completely. Heating of thermodynamically incompatible biopolymers, as the microstructure
above 30 °C causes a loss of droplet stability; the rate-limiting step for develops via spinodal decomposition, the added colloidal particles
coalescence is the melting of the Pickering crystals around the (polystyrene latex ∼300 nm) were observed to adsorb at the newly
dispersed aqueous droplets [35]. In a concentrated O/W emulsion formed liquid–liquid interface. Fig. 2 shows images of the phase-
containing interfacial fat crystals, thermal processing can lead to separated system (gelatin/starch/sugars/water/latex) held at 40 °C for
gelation of the system due to partial coalecence or a jamming (a) 16 min and (b) 16 h. Regions completely depleted of the colloidal
mechanism [36]. particles appear black in these images. We have observed [53•] that the
The important relationship between the wettability of food colloid particles are predominantly concentrated at the liquid–liquid interface
particles and the type of particle-stabilized emulsion produced has and in the regions of gelatin-rich phase. In the early stages of
been demonstrated experimentally by Paunov et al. [37••]. In this work observation (Fig. 2(a)) the phase boundaries appear fairly smooth, as
the contact angles were determined by a novel gel trapping technique
[38,39]. This involved spreading the particles at the oil–water (or air–
water) interface, and then gelling the aqueous sub-phase with a non-
adsorbing hydrocolloid (gellan gum) to fix the particles in their
original positions at the liquid interface. Subsequent moulding of the
gel surface with a curable silicone elastomer (PMDS) allowed the
replica to be imaged by SEM, so determining the particle locations
with respect to the original liquid interface. The use of this semi-
quantitative technique was demonstrated [37••] for three kinds of
contrasting particles: triglyceride crystals (10–30 μm), calcium car-
bonate particles coated with stearic acid (1–10 μm), soy protein
aggregates (1–30 μm). The fat crystals gave large contact angles at the
oil–water interface (θ ∼ 120°); W/O emulsions were stabilized
preferentially for all volume fractions of oil (n-decane). The coated
calcium carbonate particles were of intermediate wettability (θ ∼ 90°);
they gave O/W emulsions at low oil volume fractions (b50 vol.%); but
these inverted to W/O emulsions at high oil content. The soy protein
particles were predominantly hydrophilic; they produced O/W
emulsions at all volume fractions.
In addition to the various particulate systems mentioned above,
Pickering emulsions can also be stabilized by many other kinds of
particles — polystyrene latices [40], clay platelets [41,42], microfibril-
lated cellulose [43], cyclodextrin–alkane precipitates [44] and microgels
[45–47•]. The study of soft microgel particles is of particular technical
interest because of their potential applications as stimuli-responsive
stabilizers, i.e., particles whose effectiveness is sensitive to environ-
mental variables such as pH and temperature. For instance, in a study of
coarse O/W emulsions stabilized by cross-linked poly(N-isopropylacry-
lamide)-co-methacrylic acid microgel particles (∼500 nm), it was
reported [47•] that the emulsions could be broken by pH reduction
and by heating well above the volume phase transition temperature.
The responsiveness of these microgels was found to be sensitive also to
the polarity of the dispersed oil phase. Based on complementary
interfacial dilatational rheology measurements, Brugger et al. [47•]
demonstrated a correlation between the viscoelasticity of the interfacial
microgel layer and the emulsion stability. The transition from highly
elastic to less elastic interface at pH ∼ 5 could be explained in terms of a
predominant contribution of the charged methacrylic acid groups to the
elasticity of the layer.
In general, when solid particles adsorb at a liquid–liquid interface at
a high packing fraction, there is a loss of interfacial mobility. This
phenomenon is known as ‘interfacial jamming’ [48•]. The process of
jamming can arrest interfacial tension-driven morphological coarsen-
ing, leading to the stabilization of individual droplets of highly non-
spherical shapes [49,50]. This phenomenon has led to the concept of a Fig. 2. Confocal microscopy images of the structure of a phase-separated mixture of
new type of soft matter called ‘bijels’ (bicontinuous interfacially jammed gelatin (7 wt.%) + oxidized starch (4 wt.%) in an aqueous medium (25.5 wt% sucrose,
emulsion gels) [48,51,52]. A bijel is prepared from a bicontinuous 31.4 wt% glucose syrup, pH = 5.2) containing polystyrene latex particles (d32 = 313 nm,
0.7 wt.%). The sample was quenched from 90 to 1 °C, held at 1 °C for 10 min, heated to
morphology into which are introduced some adsorbing particles. The 40 °C at 6 °C min− 1, and observed at 40 °C after (a) 16 min and (b) 16 h. Dark regions
interfacial tension drives structural coarsening, and there is a gradual are depleted of particles. Starch-rich domains are identified by the symbol S.
reduction in the total interfacial area. During this process the (Reproduced from Firoozmand et al. [53•] with permission.)
44 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49

found in the absence of particles. But with increased particle mixed adsorbed layer. Uncharged nanoparticles were found to coexist
concentration and ageing time (Fig. 2(b)), the extent of wrinkling of with the adsorbed surfactant at low SDS concentrations, but these
the particle-laden liquid–liquid interface becomes very pronounced. were competitively displaced from the interface at high surfactant
The developing morphological behaviour was interpreted [53•] as being concentrations. Competitive displacement was not observed, howev-
indicative of an interfacial region that is substantially elastic (solid-like) er, in the equivalent MD simulations involving charged nanoparticles.
in character. Fig. 3 shows a snapshot of the simulated structure of the mixed layer
In emulsions prepared as encapsulation vehicles for delivery of of SDS + nanoparticles [59•]. The hydrophobic tails of the SDS
drugs or nutrients, one may ask how the release rate of the molecules are seen to be attracted to the adsorbed nanoparticles
encapsulant is affected by the presence of a solid particle layer at and also to the bulk oil phase. A general conclusion drawn from these
the oil–water interface, as opposed to, say, a surfactant-based simulations was that the interfacial tension and interfacial thickness
stabilizing layer. Experiments designed to answer this question were significantly influenced by the presence of the charged
were recently designed by Chevalier et al. [54,55] for the cases of surfactants, but not by the presence of the (very small) nanoparticles.
the topical delivery of a hydrophilic compound (caffeine) from W/O
emulsions and a lipophilic compound (retinol) from O/W emulsions. 3. Foams
Comparison was made between Pickering emulsions (50 vol.% oil)
stabilized by silica particles and the equivalent surfactant-stabilized The stabilization of bubbles and foams by particles has been
emulsions of similar average droplet size (∼ 10 μm for W/O, ∼ 3 μm for receiving great attention from researchers in recent years [2,7–10]. A
O/W). The authors found that caffeine absorption by the skin was strong motivation for this activity is the general recognition that long-
substantially faster from silica particle-stabilized W/O emulsions than term stability is much more difficult to achieve for aerated systems
from surfactant-stabilized W/O emulsions [54], and also that retinol than it is for emulsions. Hence the incorporation of colloidal particles
penetrated further into the skin when delivered from Pickering O/W as a possible way of enhancing foam stability is a matter of great
emulsions than from polysorbate-based O/W emulsions [55]. In technological and commercial significance. This is especially impor-
addition to differences in diffusive permeation rates through the tant for food systems because the presence of an aerated structure
differing structures of the two kinds of layers, it was postulated that provides the essential textural characteristics of highly popular foods
the rigidity of the multilayer silica shell around the droplets was an such as ice-cream [60]. Furthermore, the incorporation of gas bubbles
additional factor influencing the stability and breakdown of the as a full or partial replacement for dispersed fat particles can play a
emulsions during the topical administration. useful role in the development of healthier products.
With continuing advances in computer hardware and software, The inherent instability of foam arises from the high free energy of
the structural and kinetic properties of systems containing nanopar- the gas–liquid interface. This constitutes the thermodynamic driving
ticles at interfaces are now accessible to numerical modelling on the force for reducing the total interfacial area through the combined
atomistic scale. Lane et al. [56•] have used a molecular dynamics (MD) processes of bubble coalescence and disproportionation (Ostwald
simulation to determine the hydrodynamic drag on two approaching ripening). Thin films in polyhedral foams are very much greater in size
soft nanoparticles in a bulk solvent, and on a single nanoparticle than those in concentrated emulsions, and so the probability of film
approaching a plane surface. Each soft nanoparticles consisted of a rupture is higher. Also bubbles are typically larger and less dense than
rigid amorphous silica core (5 nm diameter) onto which short poly oil droplets, and so gravity creaming is much faster for dispersions of
(oxyethylene) oligomers were grafted. The authors found [56•] that gas bubbles than for the corresponding O/W emulsions. Even more
the simulated particle–particle and particle–surface forces were important, however, is the fact that the appreciable solubility of the
purely repulsive, without the oscillations normally characteristic of gas in the aqueous phase of a foam leads to a steady diffusive mass
uncoated hard spheres. transport between bubbles of different sizes under the influence of
Computer simulation of the structuring of a nanoparticle mono- local Laplace pressure gradients. Unless bubbles are embedded in a
layer at a fluid interface was first demonstrated using the technique of solid matrix, as in ice-cream [60], or surrounded by a rigid shell-like
Brownian dynamics [57]. Subsequently, using molecular dynamics, adsorbed layer, this relentless process of disproportionation inevita-
Luo et al. [58] have simulated the behaviour of small hydrocarbon-like bly leads to the loss of all but the largest bubbles in the system [2,7].
nanoparticles (1.2 nm diameter) at the water–trichloroethylene Numerous experimental studies [61,62,63•,64•,65,66,67•,68,69,
interface, followed by a recent pioneering investigation [59•] of the 70•,71–76] have demonstrated that the formation of a close-packed
competitive adsorption and cooperative interactions in adsorbing particle layer at the gas–liquid interface generates a sort of ‘colloidal
mixtures of nanoparticles + anionic surfactant molecules (SDS). armour’ that can inhibit, or even stop altogether, the destabilizing
During this simulation, clusters of nanoparticle–surfactant complex processes of bubble coalescence and coarsening. It has been established
were observed to diffuse towards the oil–water interface to form the [61–63,66,72] that individual particle-stabilized bubbles can remain

Fig. 3. Structure of adsorbed layer of negatively charged nanoparticles + anionic surfactant molecules (SDS) at the water–trichloroethylene (TCE) interface from molecular dynamics
simulation. Nanoparticles are represented as red spheres, and hydrophobic tails of SDS molecules are represented as blue chains. Sodium ions are in green. (Reproduced from Luo et
al. [59••] with permission.)
 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49 45

stable against disproportionation for days or weeks, as compared with showing the hierarchical features of a foam stabilized by a mixture of
the equivalent protein-stabilized bubbles which typically collapse silica particles+ surfactant molecules. On the macroscopic scale, this
within an hour or so [63,72]. The formal condition that a bubble is high-volume foam had a creamy white appearance (Fig. 4(a)); it pos-
stable against coarsening is the Gibbs stability criterion, E N γ/2, where E sessed a yield stress, and so the material could support its own weight
is the surface dilatational elasticity and γ is the surface tension [74]. As under gravity, rather like a dollop of whipped dairy cream. Confocal
with Pickering emulsions, the effectiveness of particle binding to the microscopy images show that the bubble sizes were in the range of 10–
interface is controlled by the contact angle [61,62,64,65]. Using a novel 50 μm (Fig. 4(b)), and that each bubble was stabilized by a dense layer of
dispersion method in ethanol/water, Binks and Horozov [64•] have colloidal particles (∼1 μm diameter) (Fig. 4(c)). The surface hydropho-
made highly stable foams using silica nanoparticles of high hydropho- bicity and optimum wettability of the stabilizing particles were con-
bicity in the absence of any added surfactant. For the routine trolled by adsorption of amphiphilic molecules of hexylamine (Fig. 4(d)).
stabilization of aqueous foams, an appropriate degree of hydrophobic The relationship between protein molecular structure and protein
particle character can be achieved in various ways, including the binding foaming behaviour is complex [79]. One contributory factor is the
of amphiphilic molecules to the surface of hydrophilic inorganic state of aggregation of the protein. The presence of colloidal particles
particles [70–72]. As with Pickering emulsions, the effectiveness of the of protein is commonly observed in microscopic images of food
particle-coated interface seems to be reinforced under conditions of protein foam systems, e.g., aerated egg albumen [80]. One may ask
particle aggregation. There is evidence in some systems [61–63,75,76] therefore whether the particle character of a food protein ingredient
for the stabilization structure involving the formation of a weak gel-like plays any major role in controlling the foaming properties. It has been
particle network throughout the aqueous phase, including the associ- tentatively suggested [1,2] that the well-known functionality of
ated particles of the ‘colloidal armour’. By analogy with the more well- ovalbumin in aerated foods could be associated with its tendency to
established phase inversion of Pickering emulsions [17], the inversion of form coagulated protein networks and surface-active stabilizing
an aqueous foam into a water-in-air powder (‘dry water’) could be particles [81]. In contrast to ovalbumin, another protein from egg
achieved [67•] by progressively increasing the particle hydrophobicity at white, lysozyme, is a poor foaming agent in its native state; but dry
constant air/water ratio, or by changing the air/water ratio at fixed heating (at 80 °C) has been shown by Desfougères et al. [82•] to
particle wettability. improve substantially its foaming properties, as well as to induce
The underlying physical mechanism for the stabilization of protein aggregation. According to these authors, the presence of
armoured bubbles by monodispersed spherical particles has been aggregates per se is not essential for the improvement of the foaming
confirmed by computer simulation [77••]. As the simulated bubble properties of lysozyme. Rather it is that the conditions (and
shrinks, depending on its radius R relative to the particle radius r, it consequently the interactions) that favour aggregation are also the
takes up a shape that is either faceted (r/R ∼ 0.1) or crumpled (r/R bb 0.1). ones that favour the formation of stable foams [82•]. In the case of the
At the same time, the surface energy and Laplace pressure decrease until milk protein β-lactoglobulin, depending on the conditions, the
a local energy minimum is reached. The metastable equilibrium state is presence of soluble or insoluble aggregates has been reported to
characterized by a mainly saddle-shaped gas–liquid interface having have a positive influence on foam stability [83–85]. Taking all this
zero mean curvature and therefore a vanishing Laplace pressure [77••]. evidence as a whole, there seems to be a general tendency for
An additional factor affecting foaming behaviour when the particle incipient loss of protein solubility to enhance interfacial viscoelastic
concentration is very high is the structuring of the particles in the bulk properties and therefore to enhance foam stability [2,79]. But in some
aqueous medium; this phenomenon was also recently demonstrated by instances, at least, the presence of aggregated protein particles (e.g.,
computer simulation [78]. by whipping or heating) may be an indirect consequence of this loss of
Foam assembly involves a high degree of synergism involving solubility rather than the direct cause of the improved foam stability.
structural building blocks on length scales spanning several orders of A unique class of proteins, possessing remarkable interfacial and
magnitude (macroscopic/microscopic/nanoscale/molecular). Fig. 4 is a foam stabilizing properties, are known as ‘hydrophobins’ [86,87••].
set of images reproduced from the work of Gonzenbach et al. [70•] These are small proteins of highly stable native structure produced by

Fig. 4. Representation of the structure of a high-volume particle-stabilized foam on various length scales: (a) macroscopic, (b) low-resolution microscopic, (c) high-resolution
microscopic, and (d) nanoscale/molecular. The confocal images in (b) and (c) were obtained following dilution of the concentrated foam (inset in (b)) containing fluorescently
labelled silica particles with hexylamine as the surfactant. (Reproduced from Gonzenbach et al. [70•] with permission.)
46 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49

filamentous fungi. They tend to self-associate into small aggregates in such an acidified aerated system through the incorporation of a suitable
aqueous media. The soluble type II hydrophobins are more surface- oil-soluble emulsifier (LACTEM) in conjunction with a sufficiently high
active than all the major food proteins (including β-casein), and they proportion of fully solid-like emulsion droplets.
have a strong tendency to self-assemble at the air–water interface to
form highly viscoelastic layers [87••]. Values of the surface shear 4. Conclusions and outlook
viscoelasticity have been found to be much higher than those of other
proteins investigated. Also, unlike other food proteins, it appears that The past few years have seen enormous growth in the amount of
the hydrophobin adsorbed layer can fully prevent bubble shrinkage fundamental and applied research relating to particle-stabilized emul-
due to disproportionation: hydrophobin-based foams have been sions and foams. Some of the newly discovered enthusiasm for this
prepared which remain stable for months or even years [88,89]. A subject is undoubtedly attributable to a desire amongst some
striking structural feature of the hydrophobin molecule is that it is researchers to join the potentially lucrative bandwagon of nanoparticle
rigid like a small solid particle; it is also amphiphilic, having a research which exists now across the whole breadth of the scientific
hydrophobic patch on one side of the molecule. A simple represen- community. A cynical observer with a long memory could therefore be
tation of an individual hydrophobin molecule might therefore be forgiven for thinking that the novelty of some recently published work is
rather like a nanosized Janus particle. In fact, some useful analogies really little more than a ‘rediscovery’ of traditional colloid science
have been drawn [90] between the interfacial behaviour of hydro- knowledge long forgotten or neglected. Whilst recognizing that this
phobin and the stabilization of interfaces by self-assembled Janus analysis does contain an element of truth, it is clear to this reviewer that
particles. a considerable amount of genuinely original and exciting information
Despite the considerable potential benefits of adsorbed particles as has been obtained recently for particle-based colloidal systems using a
stabilizing structural units in aerated systems, it is important to variety of modern instrumental methods, supported by powerful insight
recognize that the presence of even a small quantity of lipid-based from modelling and computer simulation.
particulate material in a polyhedral foam can be detrimental to The proportion of this research that is directly applicable to foods
stability [1•]. For a hydrophobic particle large enough to touch both is necessarily small. Nevertheless, it appears that the special
surfaces of the thin liquid film between a pair of bubbles, the Laplace properties of particle-laden interfaces do have much to offer in
pressure in the film adjacent to the extraneous particle may become developing the science of food structuring [101,102] and in formu-
positive. This induces liquid flow away from the particle, causing lating novel structures for improved health and wellness [103–109]. A
liquid to break contact with the particle, leading to film rupture. key challenge for the industry is to produce nanoparticles and
Another type of contaminating particle is one that spreads its contents microparticles that are both effective and also acceptable for use in
at the air–water interface. The nearby film liquid is made to move in food products on the commercial scale. Within the voluminous
the same direction as the spreading particulate material, which literature on nanoparticles, it is apparent that there are still relatively
induces a local thinning of the film, so enhancing the probability of few studies involving food-grade ingredients. There is certainly
rupture [1• ]. Both of these mechanisms are involved in the increasing interest in bionanoparticle systems, although some kinds
destabilization of aqueous food foams by fatty particles. of systems studied so far, such as virus particles [110,111], would
High density arrangements of particles can provide long-term seem to be unsuitable for food applications.
stability to aerated systems even in the absence of particle adsorption, Natural biopolymer structural assemblies are obviously attractive as
especially if the closely packed particles (e.g., sugar granules) form a (nano)particle building blocks. Polysaccharides such as starch and
continuous network structure [91]. In foods, an important class of cellulose represent a readily accessible and inexpensive source of
particle-stabilized system is the whipped dairy-based emulsion in particulate material for potential food use. And, indeed, after appropri-
which gas cells are stabilized by a network of partly coalesced fat ate hydrophobic modification, these materials can be highly effective in
globules. Destabilization of the O/W emulsion can be sensitively affected the stabilization of emulsions and foams [112,113]. However, protein-
by shearing conditions [92], the presence of emulsifiers [93,94], and the based particles probably have greater flexibility as stabilizing agents due
crystallization state of the fat [34,94]. During the whipping of natural to their natural amphiphilicity and surface activity, as well as their
cream (fat content ∼35 wt.%), the partially crystalline fat globules potential as multifunctional nanoscale delivery vehicles for neutraceu-
adhere to bubble surfaces, and subsequently they become clumped ticals [114]. The most intensively investigated of all the natural colloidal
together by a process of surface-mediated partial coalescence [95]. This particles is the casein micelle [115]. As the main stabilizing entity of
process continues until a three-dimensional network of clumped fat homogenized and recombined milk, the casein aggregate can properly
globules is built up, which holds and stabilizes the incorporated bubbles, be considered as the prototype structure for building milk protein-based
and consequently provides the desired texture and mechanical strength nanoparticles — reassembled from a mixture of sodium caseinate and
(yield stress) to the final whipped cream (typical gas-to-liquid volume calcium salts [116], or modified from the native micellar form by
ratio ∼120%). internal enzymatic cross-linking [117]. Also possessing favourable
Acid-induced aggregation of protein-coated emulsion droplets in the attributes for stabilizing emulsions and foams are whey proteins (and
vicinity of the air–water interface has been shown [96•] to have a other globular food proteins); it is now well recognized that, depending
substantial positive influence on the coalescence stability of individual on the conditions, individual whey proteins can self-assemble into
gas bubbles subjected to a sudden pressure drop. This stabilizing effect various kinds of assorted colloidal aggregates, including fibrils [118,119]
was observed to be closely correlated with corresponding changes in the and nanotubes [120]. There are also reports of more exotic bionano-
pH-dependent surface shear rheology. However, the same study also particle emulsifiers, such as ferritin, a 24-subunit iron storage protein
showed [96•] that the influence of pH and aggregated emulsion droplets composed of a hollow shell-like structure (with inner and outer
on bubble disproportionation rate and surface dilatational properties diameters of 7 and 12 nm, respectively) [121]. Other novel possibilities
was much less significant. In a separate project involving emulsified with specific functionalities could emerge from the design of nanopar-
systems of moderately high oil content (30 vol.%), it was found [97–99] ticle–protein conjugates [122].
that aerated cream-like foams could be stabilized in the complete Finally, in the opinion of this reviewer, one of most promising ways
absence of partial coalescence by whipping a caseinate-stabilized of making nanoscale structures for stabilization of food colloids is
emulsion under pH-lowering conditions which induced the emulsion through the exploitation of associative protein–polysaccharide inter-
droplets to become aggregated at the bubble surfaces. Moreover, it was actions [18,123–125•]. In considering the mechanistic and structural
subsequently demonstrated [100•] that rheological behaviour similar to consequences of protein–polysaccharide interactions in aqueous
that of traditional whipped cream (overrun 120%) could be achieved in media, there is a subtle overlap between effects of interactions on
 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49 47

the macromolecular scale and those on the colloidal scale. The [17] Binks BP. Particles as surfactants—similarities and differences. Curr Opin Colloid
Interface Sci 2003;7:21–41.
molecular forces involved may be physical (electrostatic, hydrogen [18] Dickinson E. Hydrocolloids as emulsifiers and emulsion stabilizers. Food
bonding, and hydrophobic) and/or covalent (transglutaminase- Hydrocoll 2009;23:1473–82.
catalysed cross-linking, Maillard reaction, etc.). Different kinds of [19]
•
Sacanna S, Kegel WK, Philipse AP. Thermodynamically stable Pickering emul-
sions. Phys Rev Lett 2007;98:158301. A mixture of oil + water + nanoparticles
stable composite nanoparticles have been prepared by Singh et al. (magnetite spheres, 11 nm) self-assembles spontaneously into a stable O/W
from mixtures of sodium caseinate + gum arabic, based on the emulsion with monodisperse droplets (30–150 nm). Apart from its opaque
alternative mechanisms of electrostatic complexation [126] or appearance, the particle-stabilized emulsion appears similar in properties to
those of a conventional transparent surfactant-based microemulsion.
enzymatic conjugation [127]. An important benefit of complexing [20] Wolf B, Lam S, Kirkland M, Frith WJ. Shear thickening of an emulsion stabilized
casein with a charged polysaccharide is the maintenance of solubility with hydrophilic silica particles. J Rheol 2007;51:465–78.
and functionality under acidic pH conditions [128,129]. Schmitt et al. [21] Frith WJ, Pichot R, Kirkland M, Wolf B. Formation, stability, and rheology of
particle stabilized emulsions: influence of multivalent cations. Ind Eng Chem Res
have reported directly on the interfacial and foaming properties of
2008;47:6434–44.
β-lactoglobulin–acacia gum complexes [130•]. The authors demon- [22] Li J, Stöver HDH. Doubly pH-responsive Pickering emulsion. Langmuir 2008;24:
strated that the electrostatic complexes formed a thick hydrated 13237–40.
network at the bubble surface which was more effective than the [23] Binks BP, Rodrigues JA, Frith WJ. Synergistic interaction in emulsions stabilized by a
mixture of silica nanoparticles and cationic surfactant. Langmuir 2007;23:3626–36.
pure protein in preventing destabilization due to disproportion- [24] Binks BP, Rodrigues JA. Enhanced stabilization of emulsions due to surfactant-
•
ation. In another example, aggregates of heat-denatured globular induced nanoparticle flocculation. Langmuir 2007;23:7436–9. Study of O/W
protein (e.g. β-lactoglobulin) have been stabilized by adsorbing emulsions containing positively charged alumina-coated silica particles +
sodium docyl sulfate. The most stable systems with respect to coalescence
charged polysaccharide (e.g. pectin) onto the protein aggregate were gel-like in consistency, with particle flocs at the oil–water interface.
surface, thereby producing so-called core–shell bionanoparticles [25] Schmitt-Rozières M, Krägel J, Grigoriev DO, Liggieri L, Miller R, Vincent-Bonnieu
[131]. For purposes of encapsulation, there is considerable activity in Antoni M. From spherical to polymorphous dispersed phase transition in water/
oil emulsions. Langmuir 2009;25:4266–70.
preparing nanogels and core–shell nanoparticles from mixtures of [26] Whitby CP, Fornasiero D, Ralston J. Effect of adding anionic surfactant on the
proteins with the cationic polysaccharide chitosan [132,133]. Some stability of Pickering emulsions. J Colloid Interface Sci 2009;329:173–81.
of these protein–polysaccharide complexes may have application in [27]
•
Eskandar NG, Simovic S, Prestidge CA. Synergistic effect of silica nanoparticles and
charged surfactants in the formation and stability of submicron oil-in-water
the stabilization of food emulsions and foams. emulsions. Phys Chem Chem Phys 2007;9:6426–34. For triglyceride O/W emulsions
stabilized by a mixture of silica nanoparticles + soybean lecithin (or oleylamine), it
References was demonstrated that incorporation of particles from the oil phase results in
smaller droplets and better long-term stability towards coalescence.
[1] Dickinson E. Interfacial particles in food emulsions and foams. In: Binks BP, Horozov [28]
•
Pichot R, Spyropoulos F, Norton IT. Mixed emulsifier stabilized emulsions:
•
TS, editors. Colloidal particles at liquid interfaces. Cambridge, UK: Cambridge investigation of the effect of monoolein and hydrophilic silica particle mixtures
University Press; 2006. p. 298–327. Overview of principles and practice concerning on the stability against coalescence. J Colloid Interface Sci 2009;329:284–91.
the wettability and interfacial properties of particles in food colloids, including fat From an analysis of experimental data on triglyceride O/W emulsions stabilized
crystals at the oil–water interface and clumped emulsion droplets at the air–water by silica nanoparticles + monoglyceride, a two-stage synergistic mechanism is
interface inferred involving the stabilizing role of monoglyceride during emulsification
[2] Murray BS, Ettelaie R. Foam stability: proteins and nanoparticles. Curr Opin and that of the particles during long-term emulsion storage.
Colloid Interface Sci 2004;9:314–20. [29]
•
Binks BP, Liu W, Rodriguez JA. Novel stabilization of emulsions via the
[3] Binks BP, Horozov TS. Colloidal particles at liquid interfaces: an introduction. In: heteroaggregation of nanoparticles. Langmuir 2008;24:4443–6. Surfactant-free
Binks BP, Horozov TS, editors. Colloidal particles at liquid interfaces. Cambridge: Pickering emulsions are prepared with a binary mixture of positively and negatively
Cambridge University Press; 2006. p. 1–73. charged particles.
[4] Lopetinsky RJG, Masliyah JH, Xu Z. Solids-stabilized emulsions: a review. In: Binks [30]
•
Madivala B, Vandebril S, Fransaer J, Vermant J. Exploiting particle shape in solid
BP, Horozov TS, editors. Colloidal particles at liquid interfaces. Cambridge: stabilized emulsions. Soft Matter 2009;5:1717–27. The strong influence of
Cambridge University Press; 2006. p. 186–224. particle shape on droplet coalescence stability and interfacial shear rheology is
[5] Böker A, He J, Emrick T, Russell TP. Self-assembly of nanoparticles at interfaces. demonstrated for emulsion systems containing stretched polystyrene latex
Soft Matter 2007;3:1231–48. particles or spindle-type haematite particles.
[6] Leal-Calderon L, Thivilliers F, Schmitt V. Structured emulsions. Curr Opin Colloid [31]
•
Hodge SM, Rousseau D. Flocculation and coalescence in water-in-oil emulsions
Interface Sci 2007;12:206–12. stabilized by paraffin wax crystals. Food Res Int 2003;36:695–702. Potentially
[7] Murray BS. Stabilization of bubbles and foams. Curr Opin Colloid Interface Sci stabilizing wax crystals may be present in the oil phase prior to low-temperature
2007;12:232–41. emulsification, or they may be produced in situ by rapid cooling following high-
[8] Horozov TS. Foams and foam films stabilized by solid particles. Curr Opin Colloid temperature emulsification.
Interface Sci 2008;13:134–40. [32] Rousseau D, Hodge SM. Stabilization of water-in-oil emulsions with continuous
[9] Hunter TN, Pugh RJ, Franks GV, Jameson GJ. The role of particles in stabilizing phase crystals. Colloids Surf A 2005;260:229–37.
foams and emulsions. Adv Colloid Interface Sci 2008;137:57–81. [33]
•
Binks BP, Rocher A. Effects of temperature on water-in-oil emulsions stabilized
[10] Vignes-Adler M, Weire D. New foams: fresh challenges and opportunities. Curr solely by wax microparticles. J Colloid Interface Sci 2009;335:94–104. In addition
Opin Colloid Interface Sci 2008;13:141–9. to the wax crystals, surface-active molecules from commercial wax samples can
[11] Leal-Calderon L, Schmitt V. Solid-stabilized emulsions. Curr Opin Colloid make a substantial additional contribution to the emulsion stability.
Interface Sci 2008;13:217–27. [34] Hodge SM, Rousseau D. Continuous-phase fat crystals strongly influence water-
[12] Tcholakova S, Denkov ND, Lips A. Comparison of solid particles, globular proteins in-oil emulsion stability. J Am Oil Chem Soc 2005;82:159–64.
•
and surfactants as emulsifiers. Phys Chem Chem Phys 2008;10:1608–27. [35] Rousseau D, Ghosh S, Park H. Comparison of the dispersed phase coalescence
Insightful and illuminating description of the stabilizing and destabilizing methods in different tablespreads. J Food Sci 2009;74:E1–7.
mechanisms of solid particles at interfaces in comparison with those of small- [36] Thivilliers F, Laurichesse E, Saadaoui H, Leal-Calderon F, Schmitt V. Thermally
molecule surfactants and proteins. induced gelling of oil-in-water emulsions comprising partially crystallized
[13] Binks BP, Clint JH, Mackenzie G, Simcock C, Whitby CP. Naturally occurring spore droplets: the impact of interfacial crystals. Langmuir 2008;24:13364–75.
particles at planar interfaces and in emulsions. Langmuir 2005;21:8161–7. [37]
••
Paunov VN, Cayre OJ, Noble PF, Stoyanov SD, Velikov KP, Golding M. Emulsions
[14] Arditty S, Schmitt V, Lequeux F, Leal-Calderon F. Interfacial properties in solid- stabilized by food colloid particles: role of particle adsorption and wettability at
stabilized emulsions. Eur Phys J B 2005;44:381–93. the liquid interface. J Colloid Interface Sci 2007;312:381–9. The influence of the
[15] Horozov TS, Binks BP. Particle-stabilized emulsions: a bilayer or a bridging monolayer? experimentally determined contact angle on the type of particle-stabilized
•
Angew Chem Int Ed 2006;45:773–6. The issue of the favoured arrangement of emulsion (O/W or W/O) is demonstrated directly for three kinds of food colloid
particles in the gap between adjacent stable droplets is explicitly addressed. particles: fat crystals, soy protein aggregates, and calcium carbonate particles
[16] Gautier F, Destribats M, Perrier-Cornet R, Dechézelles J-F, Giermanska J, Héroguez coated with stearic acid.
•
V, et al. Pickering emulsions with stimulable particles: from highly- to weakly- [38]
•
Paunov VN. Novel method for determining three-phase contact angle of colloidal
covered interfaces. Phys Chem Chem Phys 2007;9:6455–62. Influences of particle particles adsorbed at air–water and oil–water interfaces. Langmuir
size and interparticle interactions on the interfacial stabilizing structure are 2003;19:7970–6. Particles are fixed at the interface by gelation with gellan
considered from a comparative experimental perspective. Depending on pH and gum, and the resulting silicone elastomer replica is imaged by SEM.
ionic strength, an emulsion may contain large droplets densely covered by [39] Cayre OJ, Paunov VN. Contact angles of colloidal silica and gold particles at air–
particles or small droplets sparsely covered. water and oil–water interfaces determined with the gel trapping technique.
Langmuir 2004;20:9594–9.
[40] He X, Ge X, Liu H, Zhou H, Zhang ZC. Self-assembly of latex particles at droplet
•
Of special interest. interface to prepare monodisperse emulsion droplets. J Colloid Interface Sci
••
Of outstanding interest. 2007;301:80–4.
48 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49

[41] Torres LG, Iturbe R, Snowden MJ, Chowdhry BZ, Leharne SA. Preparation of [69] Subramaniam AB, Abkarian M, Mahadevan L, Stone HA. Mechanics of interfacial
o/w emulsions stabilized by solid particles and their characterization by oscillatory composite materials. Langmuir 2006;22:10204–8.
rheology. Colloids Surf A 2007;302:439–48. [70]
•
Gonzenbach UT, Studart AR, Tervoort E, Gauckler LJ. Ultrastable particle-
[42] Nonomura Y, Kobayashi N. Phase inversion of the Pickering emulsions stabilized stabilized foams. Angew Chem Int Ed 2006;45:3526–30. Clear concise demon-
by plate-shaped clay particles. J Colloid Interface Sci 2009;330:463–6. stration of the hierarchical structural features of a foam stabilized by a mixture of
[43] Andresen M, Stenius P. Water-in-oil emulsions stabilized by hydrophobized particles and short amphiphilic molecules.
microfibrillated cellulose. J Dispers Sci Technol 2007;28:837–44. [71] Gonzenbach UT, Studart AR, Tervoort E, Gauckler LJ. Stabilization of foams with
[44] Inoue M, Hashizaki K, Taguchi H, Saito Y. Preparation and characterization of n- inorganic colloidal particles. Langmuir 2006;22:10983–8.
alkane/water emulsion stabilized by cyclodextrin. J Oleo Sci 2009;58:85–90. [72] Kostakis T, Ettelaie R, Murray BS. Enhancement of stability of bubbles to
[45] Ngai T, Auweter H, Behrens SH. Environmental responsiveness of microgel disproportionation using hydrophilic silica particles mixed with surfactants or
particles and particle-stabilized emulsions. Macromolecules 2006;39:8171–7. proteins. In: Dickinson E, Leser ME, editors. Food colloids: self-assembly and
[46] Brugger B, Richtering W. Emulsions stabilized by stimuli-sensitive poly(N- material science. Cambridge, UK: Royal Society of Chemistry; 2007. p. 357–68.
isopropylacrylamide)-co-methacrylic acid polymers: microgels versus low [73] Safouane M, Langevin D, Binks BP. Effect of particle hydrophobicity on the properties
molecular weight polymers. Langmuir 2008;24:7769–77. of silica particle layers at the air–water interface. Langmuir 2007;23:11546–53.
[47]
•
Brugger B, Rosen BA, Richtering W. Microgels as stimuli-responsive stabilizers for [74] Cervantes-Martinez A, Rio E, Delon G, Saint-Jalmes A, Langevin D, Binks BP. On the
emulsions. Langmuir 2008;24:12202–8. As well as their sensitivity to stimuli origin of the remarkable stability of aqueous foams stabilized by nanoparticles: link
such as pH and temperature, it appears that another advantage of microgel with microscopic surface properties. Soft Matter 2008;4:1531–5.
particles is that they are relatively easy to synthesize, i.e., compared with surface- [75] Binks BP, Kirkland M, Rodrigues JA. Origin of stabilization of aqueous foams in
modified latices or inorganic/polymer hybrid particles. nanoparticle–surfactant mixtures. Soft Matter 2008;4:2373–82.
[48]
•
Stratford K, Adhikari R, Pagonabarraga I, Desplat J-C, Cates ME. Colloidal jamming [76] Zhang S, Sun D, Dong X, Li C, Xu J. Aqueous foams stabilized with particles and
at interfaces: a route to fluid bicontinuous gels. Science 2005;309:2198–201. nonionic surfactants. Colloids Surf A 2008;324:1–8.
Based on results from a computer simulation model, the concept of a new class of [77]
••
Abkarian M, Subramaniam AB, Kim S-H, Larsen R, Yang S-M, Stone HA.
soft solids called ‘bijels’ is proposed involving particle-induced arrest of the Dissolution arrest and stability of armored bubbles. Phys Rev Lett
bicontinuous interface in mixed system exhibiting spinodal decomposition. 2007;99:188301. Explanation of the stability mechanism of ‘armoured bubbles’.
[49] Bon SAF, Mookhoek SD, Colver PJ, Fischer HR, van der Zwaag S. Route to stable Computer simulation of a shrinking particle-covered bubble reveals that the
non-spherical emulsion droplets. Eur Polym J 2007;43:4839–42. metastable equilibrium state is characterized by a saddle-shaped gas–liquid
[50] Cheng H-L, Velankar SS. Controlled jamming of particle-laden interfaces using a interface having zero mean curvature and vanishing Laplace pressure.
spinning drop tensiometer. Langmuir 2009;25:4412–20. [78] Vijayaraghavan K, Nikolov A, Wasan D, Henderson D. Foamability of liquid
[51]
•
Herzig EM, White KA, Schofield AB, Poon WCK, Clegg PS. Bicontinuous emulsions particle suspensions: a modelling study. Ind Eng Chem Res 2009;48:8180–5.
stabilized solely by colloidal particles. Nat Mater 2007;6:966–71. Experimental [79] Damodaran S. Protein stabilization of emulsions and foams. J Food Sci 2005;70:
verification of the ‘bijel’ concept for silica particles (580 nm) in the 2,6-lutidine + R54–66.
water system which exhibits a lower critical solution temperature at 34 °C. [80] Lau CK, Dickinson E. Instability and structural change in an aerated system
[52]
••
Clegg PS. Fluid–bicontinuous gels stabilized by interfacial colloids: low and high containing egg albumen and invert sugar. Food Hydrocoll 2005;19:111–21.
molecular weight fluids. J Phys Condens Matter 2008;20:113101. Comprehensive [81] Croguennec T, Renault A, Beaufils S, Dubois J-J, Pezennec S. Interfacial properties
and accessible review of the emerging field of ‘bijels’ including extensive of heat-treated ovalbumin. J Colloid Interface Sci 2007;315:627–36.
bibliography and some very nice microscopic images. [82]
•
Desfougères Y, Lechevalier V, Pezennec S, Artzner F, Nau F. Dry-heating makes hen
[53]
•
Firoozmand H, Murray BS, Dickinson E. Interfacial structuring in a phase-separated egg white lysozyme an efficient foaming agent and enables its bulk aggregation. J
mixed biopolymer solution containing colloidal particles. Langmuir 2009;25: Agric Food Chem 2008;56:5120–8. While the presence of protein aggregates
1300–5. correlates with greater foaming efficiency, it is argued here that the protein particles
[54] Frelichowska J, Bolzinger M-A, Valour J-P, Mouaziz H, Pelletier J, Chevalier Y. themselves are not mechanistically responsible for the improved stability.
Pickering w/o emulsions: drug release and topical delivery. Int J Pharm 2009;368: [83] Unterhaslberger G, Schmitt C, Shojaei-Rami S, Sanchez C. Lactoglobulin
7–15. aggregates from heating with charged cosolutes: formation, characterization
[55] Frelichowska J, Bolzinger M-A, Pelletier J, Valour J-P, Chevalier Y. Topical delivery and foaming. In: Dickinson E, Leser ME, editors. Food colloids: self-assembly and
of lipophilic drugs from o/w Pickering emulsions. Int J Pharm 2009;371:56–63. material science. Cambridge, UK: Royal Society of Chemistry; 2007. p. 177–94.
[56]
•
Lane JMD, Ismail AE, Chandross M, Lorenz CD, Grest GS. Forces between [84] Rullier B, Novales B, Axelos MAV. Effect of protein aggregates on foaming of
functionalized silica nanoparticles in solution. Phys Rev E 2009;79:050501. Fully lactoglobulin. Colloids Surf A 2008;330:96–102.
atomistic molecular dynamics simulations of interaction between nanoparticles [85] Rullier B, Axelos MAV, Langevin D, Novales B. Lactoglobulin aggregates in foam
coated with polyethylene oxide oligomers. Functional coating supresses force films: correlation between foam films and foaming properties. J Colloid Interface
oscillations typically observed for rigid spheres in an explicit solvent. Sci 2009;336:750–5.
[57] Pugnaloni LA, Ettelaie R, Dickinson E. Computer simulation of the microstructure [86] Linder MB. Hydrophobins: proteins that self-assemble at interfaces. Curr Opin
of a nanoparticle monolayer formed under interfacial compression. Langmuir Colloid Interface Sci 2009;14:356–63.
2004;20:6096–9. [87]
••
Cox AR, Cagnol F, Russell AB, Izzard MJ. Surface properties of class II hydrophobins
[58] Luo MX, Mazyar OA, Zhu Q, Vaughn MW, Hase WL, Dai LL. Molecular dynamics from Trichoderma reesei and influence on bubble stability. Langmuir
simulation of nanoparticle self-assembly at a liquid–liquid interface. Langmuir 2007;23:7995–8002. Convincing measurements of the surface activity and surface
2006;22:6385–90. rheology of the unique protein hydrophobin, explaining its special ability to form an
[59]
••
Luo MX, Song Y, Dai LL. Heterogeneous or competitive self-assembly of surfactants elastic layer at the air–water interface with capability of resisting bubble shrinkage
and nanoparticles at liquid–liquid interfaces. Mol Simul 2009;10–11:773–84. due to disproportionation.
Pioneering computer simulation of the dynamics of the adsorption of a mixture of [88]
•
Cox AR, Aldred DL, Russell AB. Exceptional stability of food foams using class II
nanoparticles and charged surfactant molecules, including explicit consideration of hydrophobin HFBII. Food Hydrocoll 2009;23:366–76. Foams and bubbles can be
the effect of nanoparticle–surfactant interactions on the adsorbed layer structure. stabilized with this protein for periods of months or even years!
[60] Goff HD, Vega C. Structure-engineering of ice-cream and foam-based foods. In: [89] Tchuenbou-Magaia FL, Norton IT, Cox PW. Hydrophobins stabilized air-filled
McClements DJ, editor. Understanding and controlling the microstructure of emulsions for the food industry. Food Hydrocoll 2009;23:1877–85.
complex foods. Cambridge, UK: Woodhead; 2007. p. 558–74. [90] Walther A, Muller AHE. Janus particles. Soft Matter 2008;4:663–8.
[61] Du Z, Bilbao-Montoya MP, Binks BP, Dickinson E, Ettelaie R, Murray BS. [91] Lau CK, Dickinson E. Stabilization of aerated sugar particle systems at high sugar
Outstanding stability of particle-stabilized bubbles. Langmuir 2003;19:3106–8. particle concentrations. Colloid Surf A 2007;301:289–300.
[62] Dickinson E, Ettelaie R, Kostakis T, Murray BS. Factors controlling the formation [92] Xu W, Nikolov A, Wasan DT. Shear-induced fat particle structure variation and the
and stability of air bubbles stabilized by partially hydrophobic silica nanopar- stability of food emulsions. 1. Effects of shear history, shear rate and temperature. J
ticles. Langmuir 2004;20:8517–25. Food Eng 2005;66:97–105.
[63]
•
Murray BS, Dickinson E, Du Z, Ettelaie R, Kostakis T, Vallet J. Disproportionation [93] Xu W, Nikolov A, Wasan DT. Shear-induced fat particle structure variation and the
kinetics of air bubbles stabilized by food proteins and nanoparticles. In: Dickinson E, stability of food emulsions. 2. Effects of surfactants, protein and fat substitutes. J
editor. Food colloids: interactions, microstructure and processing. Cambridge, UK: Food Eng 2005;66:107–16.
Royal Society of Chemistry; 2005. p. 259–72. Direct experimental demonstration of [94] Norton IT, Syropoulos F, Cox PW. Effect of emulsifiers and fat crystals on shear-
the bubble-stabilizing ability of nanoparticles as compared to food proteins. induced droplet break-up, coalescence and phase inversion. Food Hydrocoll 2009;23:
[64]
•
Binks BP, Horozov TS. Aqueous foams stabilized solely by silica nanoparticles. 1521–6.
Angew Chem Int Ed 2005;44:3722–5. Using a novel dispersion technique in [95] Hotrum NE. Cohen Stuart MA, van Vliet T, van Aken GA. Proposing a relationship
ethanol/water, it is demonstrated that silica nanoparticles of enhanced hydropho- between the spreading coefficient and the whipping time of cream. In: Dickinson
bicity can stabilize foams without added surfactant. E, editor. Food colloids: interactions, microstructure and processing. Cambridge,
[65] Kostakis T, Ettelaie R, Murray BS. Effect of high salt concentration on the UK: Royal Society of Chemistry; 2005. p. 217–325.
stabilization of bubbles by silica particles. Langmuir 2006;22:1273–80. [96]
•
Murray BS, Dickinson E, Wang Y. Bubble stability in the presence of oil-in-water
[66] Subramaniam AB, Mejean C, Abkarian M, Stone HA. Microstructure, morphology, and emulsion droplets: influence of surface shear versus dilatational rheology. Food
lifetime of armoured bubbles exposed to surfactants. Langmuir 2006;22:5986–90. Hydrocoll 2009;23:1198–208. In the presence of milk proteins, bubble
[67]
•
Binks BP, Murakami R. Phase inversion of particle-stabilized materials from coalescence following pressure drop and stability to shrinkage under quiescent
foams to dry water. Nat Mater 2006;5:865–9. Experimental demonstration of the conditions are investigated as function of pH and addition of emulsion droplets.
‘catastrophic’ inversion of an aqueous foam into a water-in-air powder. Aggregated emulsion droplets reduce the extent of coalescence, but do not affect
[68] Fujii S, Ryan AJ, Armes SP. Long-range structural order, Moiré patterns, and bubble shrinkage rates. Bubble coalescence stability as a function of pH correlates
iridescence in latex-stabilized foams. J Am Chem Soc 2006;128:7882–6. well with the surface shear viscosity.
 E. Dickinson / Current Opinion in Colloid & Interface Science 15 (2010) 40–49 49

[97] Dickinson E. Structure formation in casein-based gels, foams and emulsions. [116] Semo E, Kesselman E, Danino D, Livney YD. Casein micelle as a natural nano-
Colloids Surf A 2006;288:3–11. capsular vehicle for nutraceuticals. Food Hydrocoll 2007;21:936–42.
[98] Allen KE, Dickinson E, Murray BS. Acidified sodium caseinate emulsion foams [117] Huppertz T, de Kruif CG. Structure and stability of nanogel particles prepared by
containing liquid fat: a comparison with whipped cream. Lebens Wiss Technol internal cross-linking of casein micelles. Int Dairy J 2008:18556–65.
2006;39:225–34. [118] Bromley EHC, Krebs MRH, Donald AM. Aggregation across the length scales in
[99] Allen KE, Dickinson E, Murray BS. Development of a model whipped cream: β-lactoglobulin. Faraday Discuss 2005;128:13–27.
effects of emulsion droplet liquid/solid character and added hydrocolloid. Food [119] Bolder SG, Hendrickx H, Sagis LMC, van der Linden E. Fibril assemblies in aqueous
Hydrocoll 2008;22:690–9. whey protein mixtures. J Agric Food Chem 2006;54:4229–34.
[100]
•
Allen KE, Dickinson E, Murray BS. Whipped cream-like textured systems based on [120] Graveland-Bikker JF, de Kruif CG. Unique milk protein based nanotubes: food and
acidified caseinate-stabilized oil-in-water emulsions. Int Dairy J 2008;18:1011–21. nanotechnology meet. Trends Food Sci Technol 2006;17:196–203.
Incorporation of food emulsifier (LACTEM) reduces the apparent fracture strain in [121] Fujii S, Aichi A, Muraoka M, Kishimoto N, Iwahori K, Nakamura Y, et al. Ferritin as
whipped systems containing either all-liquid or all-solid droplets, facilitating the a bionano-particulate emulsifier. J Colloid Interface Sci 2009;338:222–8.
preparation of an aerated acidified emulsion system with overrun and rheology [122] Aubin-Tam M-E, Hamad-Schifferli K. Structure and function of nanoparticle–
similar to traditional whipped cream. protein conjugates. Biomed Mater 2008;3:034001.
[101] Aguilera JM. Food product engineering: building the right structures. J Sci Food [123] Turgeon SL, Schmitt C, Sanchez C. Protein–polysaccharide complexes and
Agric 2006;86:1147–55. conjugates. Curr Opin Colloid Interface Sci 2007;12:166–78.
[102] van der Sman RGM, van der Groot AJ. The science of food structuring. Soft Matter [124] Dickinson E. Interfacial structure and stability of food emulsions as affected by
2009;5:501–10. protein–polysaccharide interactions. Soft Matter 2008;4:932–42.
[103] Aguilera JM. Food microstructure affects the bioavailability of several nutrients. J [125]
•
Schmitt C, Aberkane L, Sanchez C. Protein–polysaccharide complexes and
Food Sci 2006;72:R21–32. coacervates. In: Phillips GO, Williams PA, editors. Handbook of hydrocolloids. 2nd
[104] Ubbink J, Kruger J. Physical approaches for the delivery of active ingredients in edn. Cambridge, UK: Woodhead; 2009. p. 420–76. Comprehensive overview of the
foods. Trends Food Sci Technol 2006;17:244–54. thermodynamics, structure and functionality of protein–polysaccharide complexes,
[105] McClements DJ, Decker EA, Park Y, Weiss J. Designing food structure to control including interfacial aspects.
stability, digestion, release and absorption of lipophilic food components. Food [126] Ye A, Flanagan J, Singh H. Formation of stable nanoparticles via electrostatic
Biophys 2008;3:219–28. complexation between sodium caseinate and gum arabic. Biopolymers 2006;82:
[106] McClements DJ, Decker EA, Park Y, Weiss J. Structural design principles for 121–33.
delivery of bioactive components in neutraceuticals and functional foods. Crit [127] Flanagan J, Singh H. Conjugation of sodium caseinate and gum arabic catalysed
Rev Food Sci Nutr 2009;49:577–606. by transglutaminase. J Agric Food Chem 2006;54:7305–10.
[107] Singh H, Ye A, Horne D. Structuring food emulsions in the gastrointestinal tract to [128] Jourdain L, Leser ME, Schmitt C, Michel M, Dickinson E. Stability of emulsions
modify lipid digestion. Prog Lipid Res 2009;48:92–100. containing sodium caseinate and dextran sulfate: relationship to complexation
[108] Augustin MA, Hemar Y. Nano- and micro-structured assemblies for encapsula- in solution. Food Hydrocoll 2008;22:647–59.
tion of food ingredients. Chem Soc Rev 2009;38:902–12. [129] Semenova MG, Belyakova LE, Polikarpov YN, Antipova AS, Dickinson E. Light
[109] Palzer S. Food structures for nutrition, health and wellness. Trends Food Sci scattering study of sodium caseinate + dextran sulfate in aqueous solution:
Technol 2009;20:194–200. relationship to emulsion stability. Food Hydrocoll 2009;23:629–39.
[110] Russell JT, Lin Y, Böker A, Su L, Carl P, Zettl H, et al. Self-assembly of cross-linking of [130]
•
Schmitt, Kolodziejczyk, Leser ME. Interfacial and foam stabilization properties of
bionanoparticles at liquid–liquid interfaces. Angew Chem Int Ed 2005;44:2420–6. β-lactoglobulin–acacia gum electrostatic complexes. In: Dickinson E, editor. Food
[111] He JB, Niu ZW, Tangirala R, Wan J-Y, We XY, Kaur G, et al. Self-assembly of colloids: interactions, microstructure and processing. Cambridge, UK: Royal Society
tobacco mosaic virus at oil–water interfaces. Langmuir 2009;25:4979–87. of Chemistry; 2005. p. 289–300. Experimental demonstration of relationship
[112] Nilsson L, Bergenståhl B. Emulsification and adsorption properties of hydrophobi- between the physico-chemical properties of food biopolymer complexes and the
cally modified potato and barley starch. J Agric Food Chem 2007;55:1469–74. stabilization of foam interfaces.
[113] Wege HA, Kim S, Paunov VN, Zhong Q, Velev OD. Long-term stabilization of [131] Santipanichwong R, Suphantharika M, Weiss J, McClements DJ. Core–shell
foams and emulsions with in-situ formed microparticles from hydrophobic biopolymer nanoparticles produced by electrostatic deposition of beet pectin
cellulose. Langmuir 2008;24:9245–53. onto heat-denatured β-lactoglobulin aggregates. J Food Sci 2008;73:N23–30.
[114] Chen L. Protein micro/nanoparticles for controlled neutraceutical delivery in [132] Chen LY, Subirade M. Chitosan–β-lactoglobulin core–shell nanoparticles as
functional foods. In: McClements DJ, Decker EA, editors. Designing functional neutraceutical carriers. Biomaterials 2005;26:6041–53.
foods. Boca Raton, FL: CRC Press; 2009. p. 572–600. [133] Yu SY, Hu JH, Pan XY, Yao P, Jiang M. Stable and pH-sensitve nanogels prepared
[115] McMahon DJ, Oommen BS. Supramolecular structure of the casein micelle. J by self-assembly of chitosan and ovalbumin. Langmuir 2006;22:2754–9.
Dairy Sci 2008;91:1709–21.