Food Hydrocolloids 23 (2009) 1473–1482

Contents lists available at ScienceDirect

Food Hydrocolloids
journal homepage: www.elsevier.com/locate/foodhyd

Hydrocolloids as emulsifiers and emulsion stabilizersq

Eric Dickinson*
Procter Department of Food Science, 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: We consider the essential molecular features of hydrocolloids having the ability to act as emulsifying
Received 6 June 2008 agents and emulsion stabilizing agents. The criteria for effectiveness in protecting newly formed droplets
Accepted 5 August 2008 against flocculation and coalescence are contrasted with the requirements to maintain long-term
stability against aggregation, creaming and Ostwald ripening. To illustrate various aspects of stability
Keywords: behaviour, comparison is made between the physico-chemical characteristics of hydrocolloid emulsi-
Hydrocolloids fying agents and those of other kinds of food emulsifying agents – surfactants, proteins and nano-
Emulsifiers
particles. Interfacial complexation between protein and polysaccharide may occur through covalent
Nanoparticles
Protein adsorption
bonding or electrostatic bonding. For the case of electrostatic protein–polysaccharide complexes, the
Protein–polysaccharide interactions interfacial nanostructure and the stabilizing properties of the adsorbed layer are dependent, amongst
other things, on the sequence of adsorption of the biopolymers to the emulsion droplet surface.
! 2008 Elsevier Ltd. All rights reserved.

1. Introduction character, gelatin is really the only protein that can be properly
categorized as a hydrocolloid. Gelatin does have some emulsifying
One of the key functional roles of food hydrocolloids is in the ability, but its more characteristic roles are as a colloid stabilizer
preparation of emulsions and in the control of emulsion shelf-life. and gelling agent.
Product applications include carbonated soft drinks (Tan, 2004), This article reviews ongoing research activity having the
ice-cream (Goff, 1997), and sauces and dressings (Sikora, Badrie, potential for providing new conceptual understanding about the
Deisingh, & Kowalski, 2008). Most hydrocolloids can act as stabi- optimum requirements for emulsification and stabilization by
lizers (stabilizing agents) of oil-in-water emulsions, but only a few hydrocolloids and the basic mechanisms involved. One active area
can act as emulsifiers (emulsifying agents). The latter functionality of current research is the stabilization of emulsions by conjugates
requires substantial surface activity at the oil–water interface, and and complexes of hydrocolloids with food proteins (Dickinson,
hence the ability to facilitate the formation and stabilization of fine 2008a). Another influence on emulsifier research in general is the
droplets during and after emulsification (Dickinson, 2003, 2004). renewed interest amongst physical scientists in emulsions (and
The most widely used polysaccharide emulsifiers in food foams) stabilized by finely dispersed particles (Aveyard, Binks, &
applications are gum arabic (Acacia senegal), modified starches, Clint, 2003; Binks & Horozov, 2006; Hunter, Pugh, Franks, &
modified celluloses, some kinds of pectin, and some gal- Jameson, 2008; Leal-Calderon & Schmitt, 2008). Active research on
actomannans (Dickinson, 2003; Garti & Reichman, 1993). The nanoparticles and microparticles at interfaces is providing a stim-
surface activity of these hydrocolloids has its molecular origin in ulus for in-depth study of interfacial self-assembly of nanoparticles
either (i) the non-polar character of chemical groups attached to (Böker, He, Emrick, & Russell, 2007) and a systematic search for the
the hydrophilic polysaccharide backbone (in hydrophobically optimum conditions promoting stabilization of droplets (and
modified starch/cellulose) or (ii) the presence of a protein compo- bubbles) by various kinds of emulsifying agents (Binks, 2003;
nent linked covalently or physically to the polysaccharide (some Tcholakova, Denlov, & Lips, 2008). At the same time, natural
gums, pectins, etc.). Protein ingredients derived from milk and eggs protein-based nanoparticles – namely casein micelles – are being
are the most commonly used food emulsifying agents; but these are promoted as ideal encapsulation vehicles for neutraceuticals
not hydrocolloids (Dickinson, 1992). Due to its unique hydrophilic (Semo, Kesselman, Danino, & Livney, 2007). Against this back-
ground, the present review attempts to assess the benefits and
implications of the trend towards biopolymer nanoparticles and
q Based on Plenary Lecture at 9th International Hydrocolloids Conference
biopolymer complexes as emulsifying and stabilizing ingredients.
(Singapore, 15–19 June 2008).
We have to recognize, of course, that an important function of
* Corresponding author. Tel.: þ44 113 343 2956; fax: þ44 113 343 2982. many hydrocolloid ingredients in oil-in-water emulsions is as
E-mail address: [email protected] a structuring/thickening/gelling agent in the aqueous medium. In

0268-005X/$ – see front matter ! 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.foodhyd.2008.08.005
1474 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482

conjunction with added ‘weighting agents’ to match the densities gel-like emulsion containing trapped ‘blobs’ of hydrocolloid-
of oil and aqueous phases (Taherian, Fustier, Britten, & Ram- structured water (Dickinson, 2006a).
aswarmy, 2008), the hydrocolloid is commonly perceived to slow
down or even prevent creaming by modifying the rheology of the 2. Physico-chemical processes involved in the making of
continuous phase. Xanthan gum is especially effective in this type emulsions
of stabilizing role. This simple rheological control mechanism is
most effective at low oil volume fractions, where individual drop- To form a fine emulsion, large deformable drops must be broken
lets are separately immobilized in an entangled biopolymer down by the vigorous application of mechanical energy (Dickinson,
network, and the small buoyancy force acting on each droplet is 1994; Walstra, 1983; Walstra & Smulders, 1997). In food processing
hardly sufficient to overcome the effective yield stress of the this can be traditionally achieved using a high-speed mixer,
surrounding weak gel-like biopolymer matrix. Theoretically, a yield a colloid mill, or a high-pressure valve homogenizer. Thermody-
stress of just 10"2 Pa is sufficient to prevent the creaming of indi- namically speaking, the process is extremely inefficient, with most
vidual dispersed droplets in the size range below w10 mm (Dick- of the power being dissipated as heat.
inson, 1988). Emulsification involves the sudden creation of a large amount of
For concentrated emulsions containing a significant amount of new liquid interface. Thermodynamics tells us that, in order to
free hydrocolloid in the aqueous phase, an alternative explanation increase the oil–water surface area by an amount DA, the required
based on polymer-induced depletion forces is now regarded as work (free energy change) is DG ¼ gDA, where g is the interfacial
more appropriate (Moschakis, Murray, & Dickinson, 2005, 2006; tension. Let us suppose that we wish to make an oil-in-water
Parker, Gunning, Ng, & Robins, 1995). At very low concentrations, emulsion of 10 vol% oil containing uniform droplets of radius 1 mm
the added hydrocolloid has a destabilizing effect on the emulsion, using an emulsifier which reduces the interfacial tension to g
since the depletion flocculation induced by the non-adsorbing w5 mN m"1. From thermodynamics, we can estimate that the
hydrocolloid causes enhanced serum separation of the emulsion. theoretical work associated with making the new interface is DG w
But at higher added hydrocolloid concentrations (still <0.1 wt% for 103 J m"3. But in practice the actual amount of work required to
the case of xanthan gum), when the depletion interactions are make such an emulsion is of the order of 106 J m"3, i.e., a thousand
stronger, creaming is inhibited due to the viscoelastic character of times larger! The reason for this gross discrepancy is that small
the interconnected regions of emulsion droplets that have become droplets have highly curved interfaces, and the breaking of larger
flocculated into a gel-like network. The system becomes kinetically droplets into smaller ones requires the rapid application of
trapped on the microscopic scale in a phase-separated state. For an a disruptive force to overcome the interfacial forces holding the
emulsion containing <0.1 wt% xanthan, the local viscosity of the larger droplet together. To disrupt a droplet of radius a requires an
oil-droplet-rich regions has been estimated to be as much as 103 external pressure gradient of magnitude Dp/a ¼ 2g/a2, where Dp is
times larger than that for the neighbouring xanthan-rich regions the Laplace pressure. This implies a pressure gradient of the order
(Moschakis et al., 2006). Moreover, the oil-droplet-rich microphase of 1010 Pa m"1 (i.e. 1 kbar cm"1). During homogenization, the fluc-
viscosity has been found to increase dramatically with xanthan tuating stress differences needed to produce such a high local
concentration. So, although the xanthan-containing phase does pressure gradient are generated from the intense laminar flow
become more viscoelastic with more xanthan present in the (shear and extensional deformations) and/or inertial effects
system, the main influence of the added hydrocolloid stabilizer on (turbulence and cavitation) (Dickinson, 1994; Walstra, 1983).
the overall rheology of the emulsion is through its effect on the oil The main role of the emulsifier is to adsorb at the surface of the
droplet network. In the presence of added hydrocolloid, the freshly formed fine droplets and so prevent them from coalescing
kinetics of phase separation (leading in the long-term to enhanced with their neighbours to form larger droplets again (see Fig. 1). For
gravity creaming and macroscopic serum separation) is controlled a fixed rate of energy dissipation during emulsification, the final
in the short/medium term by the rheological behaviour of the droplet-size distribution is determined by the time taken for the
interconnected oil droplet regions. That is, the gravitationally interface to be covered with emulsifier, as compared with
unstable liquid-like emulsion has become transformed into a stable the average time interval between droplet collisions. When the

COALESCENCE
ADSORPTION

FLOCCULATION
DISRUPTION

STABILIZATION

Fig. 1. Illustration of main physico-chemical processes involved in making of emulsions. Stabilization of fine droplets requires mechanical disruption of coarse droplets accom-
panied by rapid effective adsorption of emulsifier at the new oil–water interface. Collision of droplets with insufficient coverage of emulsifier leads to coalescence and/or
flocculation.
 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482 1475

emulsifier adsorbs too slowly, or is present at too low a concentra- required to produce the minimum mean droplet size (maximum
tion, most of the individual droplets formed during the intense surface area per unit volume of oil). Fig. 2A shows data for trigly-
energy dissipation of emulsification are not retained in the final ceride oil-in-water emulsions (15 wt% oil, pH ¼ 3.0) taken from the
emulsion. This may be due to breakage of the thin film between recent careful study of Nakauma et al. (2008). The surface–volume
colliding droplets (coalescence) or sharing of the adsorbed layer mean diameter d3,2 is plotted against emulsifier concentration for
between two droplets (bridging flocculation). The latter phenom- three different hydrocolloids: gum arabic (GA), soybean soluble
enon is prevalent in concentrated emulsions (e.g., homogenized polysaccharide (SSPS), and sugar beet pectin (SBP). It was reported
cream) which have a relatively low emulsifier/oil ratio, and in less (Nakauma et al., 2008) that the d3,2 values remained constant and
concentrated systems containing mixed polymeric emulsifiers of essentially independent of hydrocolloid concentration above 1.5%,
different surface activity (Dickinson & Galazka, 1991a). For systems 4% and 10% for SBP (0.55 mm), SSPS (0.66 mm) and GA (0.82 mm),
of high emulsifier/oil ratio being homogenized in efficient equip- respectively. Hence, the relative abilities of these hydrocolloids as
ment, the droplets produced are non-flocculated and polydisperse, emulsifying agents at pH ¼ 3 were found to lie in the order
and the mean droplet size is not so dependent on the emulsifier SBP > SSPS > GA.
concentration, but rather is mainly controlled by the hydrodynamic In deciding whether to choose a hydrocolloid or a protein
processes of droplet disruption (Jafari, Assadpoor, He, & Bhandari, ingredient as the primary emulsifying agent in any food applica-
2008; Taisne, Walstra, & Cabane, 1996). tion, the manufacturer must consider the nature of the environ-
Under the turbulent flow conditions of high-pressure homoge- mental conditions to which the system will be subjected. These
nization, the transport to the interface is dominated by convection. conditions include factors such as temperature, pH, ionic strength,
This is different from diffusion under quiescent conditions which calcium ion content, and so on. When conditions are favourable,
determines the mass transport during the laboratory measurement protein emulsifiers tend to be more efficient. Proteins have greater
of time-dependent interfacial tension. In diffusive mass transport,
the most rapidly adsorbing species are low-molecular-weight
surfactants and individual protein molecules, because these have
relatively high diffusion coefficients. In convective mass transport,
however, the most rapidly adsorbing species are colloidal particles
(e.g., casein micelles) and large macromolecules (e.g., hydro-
phobically modified starch) (Nilsson & Bergenståhl, 2007a; Nilsson
et al., 2007). With particulate and aggregated macromolecular
species adsorbing to the oil–water interface, the amount of emul-
sifying agent required to saturate the surface can be much higher
than for small-molecule emulsifiers or soluble proteins.
An effective emulsifier is therefore one that (i) rapidly reduces
the interfacial tension at the freshly formed oil–water interface, (ii)
binds strongly to the interface once adsorbed, and (iii) protects the
newly formed droplets against flocculation or coalescence. This
protection against immediate recoalescence occurs in the first place
via dynamic surface tension effects (the Gibbs–Marangoni mecha-
nism) and later via repulsive colloidal interactions (electrostatic
and steric stabilization mechanisms) (Dickinson, 1992). With
polymeric emulsifiers at relatively low concentrations, the kinetics
of emulsifier adsorption is commonly the rate-determining factor.
In many food situations, the kinetics of adsorption and the
processes of interface stabilization are complicated by the fact that
the emulsifying species are polydisperse in size and in chemical
composition. During emulsification with protein as the main
emulsifier, the presence of even a small quantity of rapidly
adsorbing surfactant can facilitate substantial reduction in the
mean droplet size (Courthaudon, Dickinson, & Dalgleish, 1991). A
rather higher concentration of a miscible cosolute (e.g., 10–20%
ethanol), whose presence lowers the surface tension of the aqueous
phase, can play a similar role (Burgaud & Dickinson, 1991). Due to
the Gibbs–Marangoni mechanism, not all the fresh oil–water
interface has to be fully saturated with emulsifier for the individual
droplets to be protected against recoalescence (Dickinson, 1994). It
should be noted, however, that the Gibbs–Marangoni effect only
operates if the emulsifying agent is present in the continuous phase
(Bancroft’s rule).

3. Relative effectiveness of various hydrocolloid emulsifying

agents

It is generally important that emulsion droplets are made as

small as possible in order to minimize gravity creaming effects Fig. 2. Comparison of properties of freshly made emulsions (15 wt% triglyceride oil,
pH ¼ 3.0) prepared under controlled hydrodynamic conditions with three different
(Dickinson, 1988). A convenient way to evaluate the relative food hydrocolloids: 6, gum arabic; ,, soybean soluble polysaccharide; B, sugar beet
effectiveness of an emulsifier under controlled hydrodynamic pectin. Plotted against the hydrocolloid emulsifier concentration are (A) mean droplet
conditions is to determine the bulk emulsifier concentration diameter d3,2 and (B) zeta potential. (With permission from Nakauma et al., 2008.)
1476 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482

binding affinities and surface activities than polysaccharide-based In experimental studies of emulsifier effectiveness, a commonly
emulsifiers, and they have lower saturation surface loads (1– reported property is the zeta potential (z) of the emulsion droplets.
2 mg m"2). These factors mean that, compared with a hydrocolloid, Fig. 2B shows some reported values of z at pH ¼ 3.0 from the work
a much lower concentration of protein emulsifier (sodium of Nakauma et al. (2008) for the same three hydrocolloids (SBP,
caseinate, b-lactoglobulin, etc.) can be used to stabilized a fine oil- SSPS and GA) as considered in Fig. 2A. According to the conven-
in-water emulsion (Dickinson, 2001). The performance problem of tional wisdom, the larger the absolute magnitude of z, the greater is
protein-stabilized emulsions, however, relates to the major the electrostatic repulsion between droplets, and therefore the
contribution of electrostatic interactions to the adsorbed layer better the stability. Thus, from the data in Fig. 2B, sugar beet pectin
structure. Combined with the low surface coverage, this makes the (SBP) should produce the most stable emulsion. But, in reality, the
emulsions susceptible to destabilization under unfavourable envi- opposite has been found to be the case (Nakauma et al., 2008): as
ronmental conditions. For instance, casein-based protein emulsions measured by the change in d3,2 on storage, emulsions made with
are highly sensitive to destabilization by acidification and calcium SBP exhibited the poorest stability. This unreliability of the zeta
ions (Dickinson, 2006b), and whey protein-based emulsions are potential as a indicator of relative emulsion stability arises in part
rather heat-sensitive (Dickinson & Parkinson, 2004). In contrast, because the classical double-layer theory used to estimate z from
the large molecular size and predominant hydrophilicity of a poly- electrophoretic mobility measurements assumes a solid charged
saccharide emulsifier allows for the formation of a thicker stabi- particle which moves relative to the electrolyte dispersion medium
lizing layer that is capable of protecting droplets against at a well-defined plane of shear (Dickinson, 1992). For a poly-
aggregation over a wide range of unfavourable conditions, such as electrolyte-coated surface, the plane of shear is ill-defined because
thermal shock treatment and the addition of calcium salts (Cha- it is dependent on the unknown hydrodynamic flow of solvent
namai & McClements, 2002). within the polymer layer, which in turn is related to the generally
Starch that has been hydrophobically modified by reaction with unknown distribution of polymer, solvent and ions within the layer.
octenyl succinate anhydride has been shown to be strongly surface- Hence, it cannot be correct to treat hydrocolloid-coated emulsion
active (Prochaska, Kedziora, Le Thanh, & Lewandowicz, 2007) and to droplets as if they are simple charged spheres. More significantly,
have excellent emulsifying and emulsion stabilizing properties though, what the experimental observations of z clearly demon-
(Chanamai & McClements, 2002; Nilsson & Bergenståhl, 2007a; strate is the overriding importance of steric repulsion to the
Taherian, Fustier, & Ramaswamy, 2007). The short octenyl succinate stabilizing properties of a hydrocolloid emulsifying agent like gum
side chains anchor the carbohydrate polymers to the oil–water arabic or modified starch.
interface, and the long amylopectin backbone protects the droplets
against flocculation by the mechanism of steric stabilization (Dick- 4. Hydrocolloids versus proteins, surfactants and
inson, 1992). Good stabilization by adsorbed polysaccharides can (nano)particles
also be achieved with various surface-active derivatives of cellulose
such as hydroxypropyl (methyl)cellulose (Wollenweber, Makievski, Hydrocolloid emulsifiers have certain general functional char-
Miller, & Daniels, 2000). Like hydrophobically modified starch, these acteristics resembling those of all the other main categories of food
macromolecules form viscous adsorbed layers at liquid–liquid emulsifying agents: proteins, surfactants and solid particles.
interfaces, and their dynamic adsorption behaviour and surface Nevertheless, being hydrophilic polymer molecules, hydrocolloids
rheological properties have been the subject of recent investigations also differ in certain respects from these other species. As an aid to
(Erni et al., 2007; Mezdour, Cuvelier, Cash, & Michon, 2007; Perez, understanding the structure/function relationship of hydrocolloids,
Carrera Sanchez, Pilosof, & Rodriguez Patino, 2008). Despite this we here compare these ingredients in more detail.
extensive research activity, however, there has been no definitive When the emulsifier concentration is low, and electrostatic
demonstration that the surface rheology of a hydrocolloid is directly interactions are largely suppressed, the emulsification behaviour is
correlated with its emulsification (or foaming) behaviour. qualitatively similar for all the various types of emulsifiers. That is,
Some studies of the interfacial properties of chemically modified the initially formed droplets are only partially covered with poly-
polysaccharides have been motivated by a commercial drive to mer/protein/surfactant/particles, and the droplets coalesce with
develop replacements for gum arabic in the emulsification of others until their surfaces become protected by a dense layer of
flavour oils for soft drinks. The emulsifying properties of gum arabic molecules/particles. Additionally, bridging flocculation may occur
are associated with a high-molecular-weight fraction representing due to sharing of droplet surfaces by adsorbed particles or macro-
less than 30% of the total hydrocolloid (Randall, Phillips, & Williams, molecules (see Fig. 1). When the emulsifier concentration is high,
1988; Ray, Bird, Iacobucci, & Clark, 1995). The protein is covalently though, the mean droplet size depends on the emulsifier type
bound to the carbohydrate in the form of a mixture of arabinoga- exclusively through the interfacial tension. The relevant property
lactan–protein complexes, each containing several highly branched here is the equilibrium (static) tension for fast adsorbing surfac-
polysaccharide units linked to a common protein core. The protein tants, but it is the dynamic surface tension for the more slowly
chain firmly anchors the complex to the oil–water interface, and adsorbing macromolecular emulsifiers, i.e., proteins and, especially,
the charged polysaccharide units attached to the protein chain hydrocolloids. For particle emulsifiers acting alone, the relevant
provide a steric barrier against droplet flocculation. Gum arabic is property is the interfacial tension of the bare oil–water interface
an extremely effective emulsifier at low pH, at high ionic strength, (Tcholakova, Denkov, & Lips, 2008).
and in the presence of beverage colouring agents. But the gum is Once an emulsion has been formed, the main factor deter-
expensive to use in practice because a rather high gum/oil ratio mining stability is the strength/range of the repulsive interactions
(w1:1) is required in order to produce fine stable emulsion droplets between pairs of closely approaching droplet surfaces (Dickinson,
(d32 << 1 mm) (McNamee, O’Riordan, & O’Sullivan, 1998; Nakauma 1992; McClements, 2005). Except at very low ionic strength, when
et al., 2008). As well as chemically modified starch/cellulose, long-ranged electrostatic repulsion is important, the minimum
several replacements for gum arabic have been proposed: certain interdroplet separation is mainly determined by the physical
kinds of pectin (Akhtar, Dickinson, Mazoyer, & Langendorff, 2002; space occupied by the species present in the adsorbed monolayer.
Siew & Williams, 2008; Williams et al., 2005), other natural gums Fig. 3 illustrates that, for these different kinds of emulsifying
(Garti, 1999; Yadav, Johnston, Hotchkiss, & Hicks, 2007; Yadav, agents, the relative thickness of thin films between pairs of closely
Parris, Johnston, & Hicks, 2008), and protein–polysaccharide approaching droplet surfaces will tend to lie in the order:
conjugates (Akhtar & Dickinson, 2007; Dickinson, 2008b). surfactants < proteins < hydrocolloids < particles.
 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482 1477

to the size of a large protein molecule (say, 5–10 nm) – can be

a
regarded as being essentially irreversibly adsorbed, so long as the
contact angle is not too low (Dickinson, 2006c).
This concept of irreversible adsorption of solid (nano)particles
at the oil–water interface has led to the belief that oil droplets
possessing a full monolayer coverage of particles can be regarded
b as having essentially indefinite stability. An alternative arrange-
ment to this two-layer stabilization (Fig. 3d) would be stabiliza-
tion by a single layer of bridging particles, where the major
portion of each particle lies in the aqueous continuous phase
(Lopetinsky, Masliyah, & Xu, 2006). This type of bridging aggre-
c gation without coalescence is also possible with some protein
emulsifiers – but definitely not with surfactants, and generally not
with hydrocolloids. With particle emulsifiers in either the bilayer
or the single-layer configurations, steric hindrance prevents
particle displacement from the interface as well as lateral
displacement within the interface. Another mechanism contrib-
d uting towards the prevention of coalescence is the stability of the
thin film of continuous phase between particle-coated droplets, as
influenced by the capillary pressure in the film and the surface
rheological properties (Lopetinsky et al., 2006). Despite these
theoretical considerations, however, the limited experimental
evidence available (Tcholakova et al., 2008) does not support
the hypothesis that particle-stabilized emulsions are more stable
Fig. 3. Schematic representation of relative thicknesses of thin films between closely
with respect to droplet–droplet coalescence, as compared with
approaching oil droplets stabilized by (a) surfactants, (b) proteins, (c) hydrocolloids typical surfactant-stabilized or protein-stabilized emulsions. This
and (d) colloidal particles. The diagrams illustrate effects of differences in sizes of the same conclusion would also appear to be applicable to a compar-
different kinds of adsorbed species, i.e., surfactant molecules (0.5–1 nm), proteins (1– ison of (nano)particles and hydrocolloids as protectors against
5 nm), hydrocolloids (5–10 nm), and particles (10 nm to several mm).
coalescence.
Another phenomenon contributing to emulsion instability is
Particles adsorbed at the surface of liquid droplets provide an Ostwald ripening. This involves smaller droplets (<1 mm) shrinking
obvious mechanical barrier against flocculation and coalescence. and disappearing at the expense of the growth of bigger droplets
The protective barrier is more effective when the particles lie (>1 mm). The thermodynamic driving force for this process is the
predominantly on the convex side of the oil–water interface, i.e., difference in chemical potentials of molecules in the smaller and
they are preferentially wetted by the continuous phase. A particle’s larger droplets (Dickinson, 1992). Mass transport takes place by
position with respect to the interface is determined by the contact diffusion between droplets. Thus, for Ostwald ripening to occur at
angle q. Fig. 4 illustrates the case of a spherical particle wetted a significant rate, the dispersed phase must have appreciable
preferentially by water (q < 90$ ). The free energy of spontaneous solubility in the continuous phase. (Because gases are highly
desorption (DGd) is proportional to [r(1 " cos q)]2, where r is the soluble in water, the mechanism is especially important for
particle radius (Dickinson, 2006c). Therefore, as long as the contact aqueous foams.) Hence, the process is significant for beverage
angle is not very close to 0$ (or 180$ ), the magnitude of DGd for an emulsions containing water-soluble flavour oils (D-limonene, etc.),
adsorbed particle of colloidal size is extremely high compared with whereas it is insignificant for dairy-type emulsions containing
the thermal energy (kT), and so the particle can be regarded as water-insoluble triglyceride oils. For a water-soluble soluble oil, the
irreversibly adsorbed. Even a relatively small nanoparticle – down most effect way to resist the ripening process is to mix it with an
insoluble oil (e.g., vegetable oil þ citrus oil in a cloudy beverage
emulsion formulation).
Depending on the type of emulsifier, there are important
γpw differences in the Ostwald ripening behaviour (Tcholakova et al.,
WATER 2008). With low-molecular-weight surfactants, the interfacial
tension is low, which reduces the thermodynamic driving force.
And because surfactant adsorption is reversible, the tension
remains essentially constant as the emulsifier is continuously
PARTICLE desorbed during droplet shrinkage (see Fig. 5a). The presence of
micelles in the aqueous phase enhances further the effective
θ solubility of the dispersed phase, which increases the mass trans-
port rate.
With a typical globular protein as emulsifier, droplet shrinkage
γow leads to a greater surface coverage and a thicker layer, since the
emulsifier resists desorption. The consequent lowering of the
OIL tension, associated with the high viscoelasticity of the adsorbed
layer, leads to a slowing down of the ripening process. But the
γpo ripening rate of the emulsion can be stopped completely in only
two ways: (a) if there are insoluble species present in the oil phase
Fig. 4. Sketch of the location of a single spherical particle at the oil–water interface.
The contact angle q is the angle formed at the junction point of particle (p), oil (o) and
(like vegetable oil added to an essential oil in a cloudy beverage
water (w). The tensions at the three separate interfaces are gpo, gpw and gow, and the emulsion), or if the (elastic) protein layer reaches a thickness
angle is defined by cos q ¼ (gpo " gpw)/gow (Young’s equation). similar to the droplet radius (Meinders & van Vliet, 2004). Over
1478 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482

a b c

↓ ↓ ↓

Fig. 5. Schematic representation of the effect of emulsifier type on the shrinkage behaviour of a small emulsion droplet as a consequence of Ostwald ripening: (a) surfactant-
stabilized, (b) protein-stabilized, and (c) particle-stabilized.

extended time-scales, a monolayer protein film is predominantly Covalent linkage of protein to polysaccharide can be achieved in
viscous, which means that there is no complete arrest of the various ways. Direct chemical means can be used to form gelatin–
ripening process: compression to high surface pressures leads to pectin hybrids linked by amide bonds by incubating a mixture of
slow protein desorption, and local interfacial stresses around small gelatin (type A) þ high-methoxyl pectin under alkaline conditions
droplets causes wrinkling of the residual protein film (see Fig. 5b). (Diftis, Pirzas, & Kiosseoglou, 2005). Alternatively, conjugation may
(In the case of a shrinking (globular) protein-stabilized gas bubble, be achieved enzymatically, e.g., using peroxidase to attach b-casein
there is complete loss of the dispersed phase, leaving just a residual to feruloylated arabinoxylan (Boeriu et al., 2004) or trans-
protein aggregate – Dickinson, Ettelaie, Murray, & Du, 2002.) When glutaminase to attach sodium caseinate to gum arabic (Flanagan &
the emulsifier is sodium casein(ate), which contains disordered Singh, 2006). But probably the most convenient method of conju-
amphiphilic protein molecules, the mechanistic behaviour can be gation is simply to heat an intimate dry mixture of the protein þ
considered as intermediate between the surfactant case and the polysaccharide (Dickinson & Galazka, 1991b; Kato, Sasaki, Furuta, &
globular protein case. Kobayashi, 1990; Oliver, Melton, & Stanley, 2006). This dry-heat-
Due to the exceptionally high desorption energy, particulate induced conjugation invariably improves protein solubility and
emulsifiers are able to prevent Ostwald ripening completely (Binks, stability under unfavourable solution conditions of low pH and
2003; Cervantes Martinez et al., 2008). The reduction in surface high ionic strength, with consequent benefits for protein emulsi-
area on droplet shrinkage produces a tightly jammed particle fication properties (Oliver et al., 2006; Dickinson, 2008a). The
monolayer (see Fig. 5c). This leads to an inward bending of the oil– growing acceptance of these Maillard-type conjugates as poten-
water interface between the particles. The creation of micro- tially valuable functional ingredients in various food applications is
configurations having zero mean curvature of fluid interface indicated by the broad range of protein þ polysaccharide combi-
between the particles (i.e., a vanishing apparent interfacial tension) nations that have been recently investigated (see Table 1).
eliminates the capillary pressure that drives the ripening process Due to the bound protein, the conjugate is much more surface-
(Tcholakova et al., 2008). Any hydrocolloid emulsifier that could active than the polysaccharide on its own; hence the conjugate is
behave like a rigid spherical nanoparticle at the oil–water interface able to achieve surface layer saturation at a much lower bulk
could also produce a tightly jammed stabilizing monolayer. But concentration. At the same time, due to the covalently bound
most hydrocolloids have ‘soft’ polymer-like structural characteris- polysaccharide, the adsorbed protein layer is protected against
tics, more resembling proteins than solid particles; hence they destabilization under unfavourable environmental conditions (e.g.,
cannot fully arrest Ostwald ripening. heating, low pH, high electrolyte concentrations, etc.). The large
size and hydrophilicity of the polysaccharide moiety generates
5. Interactions of hydrocolloids with adsorbed proteins long-range steric repulsion between emulsion droplets surfaces.
Theoretical calculations have recently shown (Ettelaie, Akinshina, &
There is increasing interest in exploiting the combined advan- Dickinson, in press) that a strong association of a specific part of
tages of proteins and hydrocolloids as functional ingredients via the a moderately large polysaccharide entity (>5–10 kDa) to an
development of protein–polysaccharide complexes as emulsifiers adsorbing protein can confer sufficient additional steric stabiliza-
and stabilizers (Dickinson, 1995, 2008a, 2008b; Guzey & McCle- tion to overcome any surface–surface attractive interaction that
ments, 2006). The protein and polysaccharide components may be might be present with the adsorbed protein alone under unfav-
joined together by either (i) covalent bonding or (ii) electrostatic ourable conditions of pH and/or ionic strength. While this interfa-
interactions. cial biopolymer structure is most efficiently realised with
 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482 1479

Table 1
Experimental studies of the characterization and functionality of Maillard-type
protein–polysaccharide conjugates published during 2004–2007

Protein Polysaccharide Reference(s)

b-Casein Dextran Mu, Pan, Yao, & Jiang, 2006
b-Lactoglobulin Chitosan Miralles, Martinez-Rodriguez,
Santiago, van de Lagemaat,
& Heras, 2007
b-Lactoglobulin Dextran Dunlap & Côté, 2005;
Jimenez-Castaño, Villamiel,
& Lopez-Fandiño, 2007;
Jimenez-Castaño, Villamiel,
Martin-Alvarez, Olano, &
Lopez-Fandiño, 2005;
Wooster & Augustin, 2006
b-Lactoglobulin Gum arabic Schmitt, Bovay, & Frossard, 2005
Ovalbumin Dextran Choi, Kim, Park, & Moon, 2005
Sodium caseinate Dextran Fechner, Knoth, Scherze,
& Muschiolik, 2007
Sodium caseinate Maltodextrin Morris, Sims, Robertson,
& Furneaux, 2004
Sodium caseinate Pectin Einhorn-Stoll, Ulbrich, Sever,
& Kunzek, 2005
Fig. 6. Influence of sodium caseinate on the time-dependent surface shear viscosity
Soybean protein Dextran Diftis, Biliaderis, & Kiosseoglou, 2005
h(t) of gelatin at the n-hexadecane–water interface (pH ¼ 7, 25 $ C). The film adsorbed
Soybean protein Porphyran Takano, Hattori, Yoshida,
from a 10"3 wt% gelatin solution had been aged for 24 h (q) or 72 h (+) prior to
Kanuma, & Takahashi, 2007
introducing 10"3 wt% caseinate into the aqueous sub-phase. The dashed line denotes
Soy protein Curdlan Fan et al., 2006
h(t) for 10"3 wt% caseinate in the absence of gelatin. Inset pictures 1 and 2 show
hydrolysate
schematic representations of the structure in the vicinity of the O/W surface (S) before
Whey protein Carboxymethyl Kita, Korlos, & Kiosseoglou, 2007
and after penetration of casein molecules (C) into the adsorbed layer of gelatin
cellulose
molecules (G). In the postulated structure after caseinate addition, region I mainly
Whey protein Dextran Wooster & Augustin, 2007
determines the surface pressure and region II dominates the surface rheology. (Taken
Whey protein Maltodextrin Akhtar & Dickinson, 2007
from Dickinson et al., 1987.)
Whey protein Pectin Einhorn-Stoll et al., 2005;
Nierynck, van der
Meeren, Bayarri Gorbe,
For use as particulate structuring agents in food colloids, it
Dierckx, & Dewettinck, 2004
would seem that there is considerable potential for the exploitation
of electrostatic protein–polysaccharide complexation in the
a permanent covalent linkage, a more common situation in food formulation of stable biopolymer-based nanoparticles by simple
emulsion systems is where protein–polysaccharide complexation mixing, e.g., whey protein þ gum arabic (Weinbreck, de Vries,
arises from non-covalent association driven by electrostatic inter- Schrooyen, & de Kruif, 2003) or sodium caseinate þ gum arabic (Ye,
actions (Dickinson, 1998). Flanagan, & Singh, 2006). Also novel protein–polysaccharide
Two of the most traditional food ingredients are casein and aggregates may be formed using static high-pressure technology to
gelatin. While both these ingredients consist of macromolecules dissociate and reassemble of native casein micelle particles in the
that are disordered, polydisperse and heterogeneous, the two kinds presence of interacting hydrocolloids such as low-methoxyl pectin
of biopolymers differ substantially in terms of hydrophilic/hydro- or i-carrageenan (Abbasi & Dickinson, 2002, 2004).
phobic balance, surface activity and intermolecular interactions. In model emulsion studies, there have been numerous cases
Furthermore, at neutral pH in the absence of calcium ions, whereas reported in the recent literature in which the rheological and
sodium caseinate forms low-viscosity mobile adsorbed layers at the stability properties can be attributable to the presence of associa-
oil–water interface, gelatin forms time-dependent gel-like stabi- tive interfacial interactions between protein and polysaccharide
lizing layers at the surface of emulsion droplets. This contrasting ingredients. Some recent examples include systems containing
behaviour has been demonstrated (Dickinson, Murray, & Stainsby, whey protein þ carboxymethylcellulose (Girard, Turgeon, & Paquin,
1985) in interfacial experiments on systems containing a binary 2002), sodium caseinate þ low-methyoxl pectin (Matia-Merino &
mixture of the two proteins. Fig. 6 shows the observed change in Dickinson, 2004), sodium caseinate þ high-methoxyl pectin
surface shear viscosity on injecting sodium caseinate below (Bonnet, Corredig, & Alexander, 2005), b-lactoglobulin þ high-
a previously adsorbed gelatin layer at the planar hydrocarbon– methoxyl pectin (Guzey, Kim, & McClements, 2004), canola
water interface (Dickinson, Murray, Murray, & Stainsby, 1987). protein þ k-carrageenan (Uruakpa & Arntfield, 2005), whey protein
While the surface viscosity was at first greatly diminished, it þ xanthan gum (Benichou, Aserin, Lutz, & Garti, 2007), and sodium
subsequently recovered to values similar to those of the original caseinate þ gellan gum (Sosa-Herrera, Berli, & Martinez-Padilla,
aged gelatin film. The longer the film had been aged prior to 2008).
addition of caseinate, the more rapid was the recovery. This Together with other factors like pH and ionic strength, the
behaviour can be interpreted in terms of the competitive mesoscopic structure of the composite interfacial layer containing
displacement of the original primary adsorbed layer of gelatin by both protein and polysaccharide has been recognized to be
the more surface-active caseinate, followed by restructuring of the dependent on the procedure used to make the emulsion (Dick-
displaced hydrocolloid as a secondary gel-like layer below the new inson, 2008b). Two alternative procedures are differentiated in
primary caseinate layer (as shown diagrammatically in the insets to Fig. 7. Method (a) is to prepare a mixed solution of the biopolymers,
Fig. 6). Though gelatin is unique in exhibiting thermoreversible and then use the resulting protein–polysaccharide complex as the
self-association behaviour through triple-helix hydrogen bonding emulsifying agent during homogenization. Method (b) is to make
interactions, a qualitatively similar kind of protein–hydrocolloid the emulsion initially with protein as emulsifying agent, mix the
organization at the oil–water interface can be expected with many washed emulsion with polysaccharide solution, and then allow the
other systems possessing favourable associative biopolymer– polysaccharide to adsorb onto the protein monolayer as a com-
biopolymer interactions. plexing secondary layer. For convenience of identification, we may
1480 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482

WATER

OIL

Fig. 7. Illustration of two alternative ways of making a composite layer composed of protein–polysaccharide complex at the oil–water interface: (a) the complex previously formed
in solution adsorbs at bare interface; (b) the polysaccharide adsorbs onto a previously formed protein monolayer.

designate these two kinds of systems as (a) ‘mixed emulsions’ and Burgaud, I., & Dickinson, E. (1991). Emulsifying effects of food macromolecules in
presence of ethanol. Journal of Food Science, 55, 875–876.
(b) ‘bilayer emulsions’ (Jourdain, Leser, Schmitt, & Dickinson,
Cervantes Martinez, A., Rio, E., Delon, G., Saint-Jalmes, A., Langevin, D., & Binks, B. P.
2008a, 2008b). Looking ahead to the future, there is good potential (2008). On the origin of the remarkable stability of aqueous foams stabilized by
for use of such mixed interfacial layers in emulsions containing nanoparticles: link with microscopic surface properties. Soft Matter, 4, 1531–
hydrocolloids in the development of delivery vehicles for nutrient 1535.
Chanamai, R., & McClements, D. J. (2002). Comparison of gum arabic, modified
encapsulation, and also in the protection of adsorbed proteins and starch, and whey protein isolate as emulsifiers: influence of pH, CaCl2 and
emulsified lipids against enzymatic breakdown during digestion temperature. Journal of Food Science, 67, 120–125.
(Dickinson, 2008b). Choi, S. J., Kim, H. J., Park, K. H., & Moon, T. W. (2005). Molecular characteristics of
ovalbumin–dextran conjugates formed through the Maillard reaction. Food
Finally, it may be noted that the bilayer concept is not limited to Chemistry, 92, 93–99.
just protein þ hydrocolloid systems: it can be applied also to the Courthaudon, J.-L., Dickinson, E., & Dalgleish, D. G. (1991). Competitive adsorption
case of electrostatic interaction between an adsorbed ionic of b-casein and non-ionic surfactants in oil-in-water emulsions. Journal of
Colloid and Interface Science, 145, 390–395.
surfactant in the primary layer and an oppositely charged hydro- Dickinson, E. (1988). The structure and stability of emulsions. In J. M. V. Blanshard, &
colloid in the secondary layer (Nilsson & Bergenståhl, 2007b). J. R. Mitchell (Eds.), Food structuredIts creation and evaluation (pp. 41–57).
Furthermore, the concept can be extended to more complex London: Butterworths.
Dickinson, E. (1992). An introduction to food colloids. Oxford: Oxford University
multilayer systems containing three (or even more) different layers Press.
(e.g., surfactant–protein–polysaccharide) (Bergenståhl, Fäldt, & Dickinson, E. (1994). Emulsions and droplet size control. In D. J. Wedlock (Ed.),
Malmsten, 1995; Guzey & McClements, 2006). Further research Controlled particle, droplet and bubble formation (pp. 191–216). Oxford: Butter-
worth–Heinemann.
activity on the properties of these mixed and bilayer formulations
Dickinson, E. (1995). Emulsion stabilization by polysaccharides and protein–poly-
can be confidently anticipated over the next few years. saccharide complexes. In A. M. Stephen (Ed.), Food polysaccharides and their
applications (pp. 501–515). New York: Marcel Dekker.
Dickinson, E. (1998). Stability and rheological implications of electrostatic milk
References protein–polysaccharide interactions. Trends in Food Science and Technology, 9,
347–354.
Abbasi, S., & Dickinson, E. (2002). High-pressure induced rheological changes of low Dickinson, E. (2001). Milk protein interfacial layers and the relationship to emulsion
methoxyl pectin þ micellar casein mixtures. Journal of Agricultural and Food stability and rheology. Colloids and Surfaces B, 20, 197–210.
Chemistry, 50, 3559–3565. Dickinson, E. (2003). Hydrocolloids at interfaces and the influence on the properties
Abbasi, S., & Dickinson, E. (2004). Gelation of i-carrageenan and micellar casein of dispersed systems. Food Hydrocolloids, 17, 25–39.
mixtures under high hydrostatic pressure. Journal of Agricultural and Food Dickinson, E. (2004). Effect of hydrocolloids on emulsion stability. In P. A. Williams,
Chemistry, 52, 1705–1714. & G. O. Phillips (Eds.), Gums and stabilisers for the food industryd12 (pp. 394–
Akhtar, M., & Dickinson, E. (2007). Whey protein–maltodextrin conjugates as 404). Cambridge, UK: Royal Society of Chemistry.
emulsifying agents: an alternative to gum arabic. Food Hydrocolloids, 21, Dickinson, E. (2006a). Colloid science of mixed ingredients. Soft Matter, 2, 642–652.
607–616. Dickinson, E. (2006b). Structure formation of casein-based gels, foams and emul-
Akhtar, M., Dickinson, E., Mazoyer, J., & Langendorff, V. (2002). Emulsion stabilizing sions. Colloids and Surfaces A, 288, 3–11.
properties of depolymerized pectin. Food Hydrocolloids, 16, 249–256. Dickinson, E. (2006c). Interfacial particles in food emulsions and foams. In
Aveyard, R., Binks, B. P., & Clint, J. H. (2003). Emulsions stabilized solely by colloidal B. P. Binks, & T. S. Horozov (Eds.), Colloidal particles at liquid interfaces (pp. 298–
particles. Advances in Colloid and Interface Science, 100–102, 503–546. 327). Cambridge, UK: Cambridge University Press.
Benichou, A., Aserin, A., Lutz, R., & Garti, N. (2007). Formation and characterization Dickinson, E. (2008a). Emulsification and emulsion stabilization with protein–
of amphiphilic conjugates of whey protein isolate (WPI)/xanthan to improve polysaccharide complexes. In P. A. Williams, & G. O. Phillips (Eds.), Gums and
surface activity. Food Hydrocolloids, 21, 379–391. stabilisers for the food industryd14 (pp. 221–232). Cambridge, UK: Royal Society
Bergenståhl, B., Fäldt, P., & Malmsten, M. (1995). Surface structures and surface-active of Chemistry.
components in food emulsions. In E. Dickinson, & D. Lorient (Eds.), Food macro- Dickinson, E. (2008b). Interfacial structure and stability of food emulsions as
molecules and colloids (pp. 201–214). Cambridge, UK: Royal Society of Chemistry. affected by protein–polysaccharide interactions. Soft Matter, 4, 932–942.
Binks, B. P. (2003). Particles as surfactants – similarities and differences. Current Dickinson, E., Ettelaie, R., Murray, B. S., & Du, Z. (2002). Kinetics of disproportion-
Opinion in Colloid and Interface Science, 7, 21–41. ation of air bubbles beneath a planar air–water interface stabilized by food
Binks, B. P., & Horozov, T. S. (Eds.). (2006). Colloidal particles at liquid interfaces. proteins. Journal of Colloid and Interface Science, 252, 202–213.
Cambridge, UK: Cambridge University Press. Dickinson, E., & Galazka, V. B. (1991a). Bridging flocculation induced by competitive
Boeriu, C. G., Oudgenoeg, G., Spekking, W. T. J., Berendsen, L. B. J. M., Vancon, L., adsorption: implications for emulsion stability. Journal of the Chemical Society,
Boumans, H., et al. (2004). Horseradish peroxidase-catalyzed cross-linking of Faraday Transactions, 87, 963–969.
feruloylated arabinoxylans with b-casein. Journal of Agricultural and Food Dickinson, E., & Galazka, V. B. (1991b). Emulsion stabilization by ionic and covalent
Chemistry, 52, 6633–6639. complexes of b-lactoglobulin with polysaccharides. Food Hydrocolloids, 4,
Böker, A., He, J., Emrick, T., & Russell, T. P. (2007). Self-assembly of nanoparticles at 281–296.
interfaces. Soft Matter, 3, 1231–1248. Dickinson, E., Murray, A., Murray, B. S., & Stainsby, G. (1987). Properties of adsorbed
Bonnet, C., Corredig, M., & Alexander, M. (2005). Stabilization of caseinate-covered layers in emulsions containing a mixture of caseinate and gelatin. In E. Dick-
oil droplets during acidification with high methoxyl pectin. Journal of Agricul- inson (Ed.), Food emulsions and foams (pp. 86–99). London: Royal Society of
tural and Food Chemistry, 53, 8600–8606. Chemistry.
 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482 1481

Dickinson, E., Murray, B. S., & Stainsby, G. (1985). Time-dependent surface viscosity Meinders, M. B. J., & van Vliet, T. (2004). The role of interfacial properties on Ostwald
of adsorbed films of casein þ gelatin at the oil–water interface. Journal of Colloid ripening in emulsions. Advances in Colloid and Interface Science, 108–109,119–126.
and Interface Science, 106, 259–262. Mezdour, S., Cuvelier, G., Cash, M. J., & Michon, C. (2007). Surface rheological
Dickinson, E., & Parkinson, E. L. (2004). Heat-induced aggregation of milk protein- properties of hydroxypropyl cellulose at air–water interface. Food Hydrocolloids,
stabilized emulsions: sensitivity to processing and composition. International 21, 776–781.
Dairy Journal, 14, 635–645. Miralles, B., Martinez-Rodriguez, A., Santiago, A., van de Lagemaat, J., & Heras, A.
Diftis, N. G., Biliaderis, C. G., & Kiosseoglou, V. D. (2005). Rheological properties and (2007). The occurrence of a Maillard-type protein–polysaccharide interaction
stability of model salad dressing emulsions prepared with a dry-heated reaction between b-lactoglobulin and chitosan. Food Chemistry, 100, 1071–1075.
soybean protein isolate–dextran mixture. Food Hydrocolloids, 19, 1025–1031. Morris, G. A., Sims, I. M., Robertson, A. J., & Furneaux, R. H. (2004). Investigation into
Diftis, N. G., Pirzas, T. A., & Kiosseoglou, V. D. (2005). Emulsifying properties of the physical and chemical properties of sodium caseinate–maltodextrin glyco-
gelatin conjugated to pectin under alkaline conditions. Journal of the Science of conjugates. Food Hydrocolloids, 18, 1007–1014.
Food and Agriculture, 85, 804–808. Moschakis, T., Murray, B. S., & Dickinson, E. (2005). Microstructural evolution of
Dunlap, C. A., & Côté, G. L. (2005). b-Lactoglobulin–dextran conjugates: effect of viscoelastic emulsions stabilized by sodium caseinate and xanthan gum. Journal
polysaccharide size on emulsion stability. Journal of Agricultural and Food of Colloid and Interface Science, 284, 714–728.
Chemistry, 53, 419–423. Moschakis, T., Murray, B. S., & Dickinson, E. (2006). Particle tracking using confocal
Einhorn-Stoll, U., Ulbrich, M., Sever, S., & Kunzek, H. (2005). Formation of milk microscopy to probe the microrheology in a phase-separating emulsion.
protein–pectin conjugates with improved emulsifying properties by controlled Langmuir, 22, 4710–4719.
dry heating. Food Hydrocolloids, 19, 329–340. Mu, M. F., Pan, X. Y., Yao, P., & Jiang, M. (2006). Acidic solution properties of b-
Erni, P., Windhab, E. J., Gunde, R., Graber, M., Pfister, B., Parker, A., et al. (2007). casein-graft-dextran copolymer prepared through Maillard reaction. Journal of
Interfacial rheology of surface-active biopolymers: Acacia senegal versus Colloid and Interface Science, 301, 98–106.
hydrophobically modified starch. Biomacromolecules, 8, 3458–3466. Nakauma, M., Funami, T., Noda, S., Ishihara, S., Al-Assaf, S., Nishinari, K., et al. (2008).
Ettelaie, R., Akinshina, A., & Dickinson, E. Mixed protein–polysaccharide interfacial Comparison of sugar beet pectin, soybean soluble polysaccharide, and gum
layers: a self consistent field calculation study. Faraday Discussions, in press. arabic as emulsifiers. 1. Effect of concentration, pH and salts on the emulsifying
DOI: 10.1039/b717199d. properties. Food Hydrocolloids, 22, 1254–1267.
Fan, J. F., Zhang, Y. Y., Szesze, T., Li, F. J., Zhou, M. Y., Saito, M., et al. (2006). Improving Nierynck, N., van der Meeren, P., Bayarri Gorbe, S., Dierckx, S., & Dewettinck, K.
functional properties of soy protein hydrolysate by conjugation with curdlan. (2004). Improved emulsion stabilizing properties of whey protein isolate by
Journal of Food Science, 71, C285–C291. conjugation with pectins. Food Hydrocolloids, 18, 949–957.
Fechner, A., Knoth, A., Scherze, I., & Muschiolik, G. (2007). Stability and release Nilsson, L., & Bergenståhl, B. (2007a). Emulsification and adsorption properties of
properties of double emulsions stabilized by caseinate–dextran conjugates. hydrophobically modified potato and barley starch. Journal of Agricultural and
Food Hydrocolloids, 21, 943–952. Food Chemistry, 55, 1469–1474.
Flanagan, J., & Singh, H. (2006). Conjugation of sodium caseinate and gum arabic Nilsson, L., & Bergenståhl, B. (2007b). Adsorption of hydrophobically modified
catalyzed by transglutaminase. Journal of Agricultural and Food Chemistry, 54, anionic starch at oppositely charged oil/water interfaces. Journal of Colloid and
7305–7310. Interface Science, 308, 508–513.
Garti, N. (1999). What can nature offer from an emulsifier point of view: trends and Nilsson, L., Osmark, P., Leeman, M., Fernadez, C., Wahlund, K.-G., & Bergenståhl, B.
progress? Colloids and Surfaces A, 152, 125–126. (2007). Adsorption of macromolecules at oil–water interfaces during emulsifi-
Garti, N., & Reichman, D. (1993). Hydrocolloids as food emulsifiers and stabilizers. cation. In E. Dickinson, & M. E. Leser (Eds.), Food colloids: self-assembly and
Food Microstructure, 12, 411–426. material science (pp. 433–448). Cambridge, UK: Royal Society of Chemistry.
Girard, M., Turgeon, S. L., & Paquin, P. (2002). Emulsifying properties of whey Oliver, C. M., Melton, L. D., & Stanley, R. A. (2006). Creating proteins with novel
protein–carboxymethylcellulose complexes. Journal of Food Science, 67, 113–119. functionality via the Maillard reaction: a review. Critical Reviews in Food Science
Goff, H. D. (1997). Colloidal aspects of ice cream – a review. International Dairy and Nutrition, 46, 337–350.
Journal, 7, 363–373. Parker, A., Gunning, P. A., Ng, K., & Robins, M. M. (1995). How does xanthan stabilize
Guzey, D., Kim, H. J., & McClements, D. J. (2004). Factors influencing the production salad dressing? Food Hydrocolloids, 9, 333–342.
of o/w emulsions stabilized by b-lactoglobulin–pectin membranes. Food Perez, O. E., Carrera Sanchez, C., Pilosof, A. M. R., & Rodriguez Patino, J. M. (2008).
Hydrocolloids, 18, 967–975. Dynamics of adsorption of hydroxypropyl methylcellulose at the air–water
Guzey, D., & McClements, D. J. (2006). Formation, stability and properties of interface. Food Hydrocolloids, 22, 387–402.
multilayer emulsions for application in the food industry. Advances in Colloid Prochaska, K., Kedziora, P., Le Thanh, J., & Lewandowicz, G. (2007). Surface activity
and Interface Science, 128–130, 227–248. of commercial food grade modified starches. Colloids and Surfaces B, 60, 187–
Hunter, T. N., Pugh, R. J., Franks, G. V., & Jameson, G. J. (2008). The role of particles in 194.
stabilizing foams and emulsions. Advances in Colloid and Interface Science, 137, Randall, R. C., Phillips, G. O., & Williams, P. A. (1988). The role of the proteinaceous
57–81. component on the emulsifying properties of gum arabic. Food Hydrocolloids, 2,
Jafari, S. M., Assadpoor, E., He, Y., & Bhandari, B. (2008). Re-coalescence of 131–140.
emulsion droplets during high-energy emulsification. Food Hydrocolloids, 22, Ray, A. K., Bird, P. B., Iacobucci, G. A., & Clark, B. C., Jr. (1995). Functionality of gum
1191–1202. arabic: fractionation, characterization and evaluation of gum fractions in citrus
Jimenez-Castaño, L., Villamiel, M., & Lopez-Fandiño, R. (2007). Glycosylation of oil emulsions and model beverages. Food Hydrocolloids, 9, 123–131.
individual whey proteins by Maillard reaction using dextran of different Schmitt, C., Bovay, C., & Frossard, P. (2005). Kinetics of formation and functional
molecular mass. Food Hydrocolloids, 21, 433–443. properties of conjugates prepared by dry-state incubation of b-lactoglobulin/
Jimenez-Castaño, L., Villamiel, M., Martin-Alvarez, P. J., Olano, A., & Lopez- acacia gum electrostatic complexes. Journal of Agricultural and Food Chemistry,
Fandiño, R. (2005). Effect of the dry-heating conditions on the glycosylation of 53, 9089–9099.
b-lactoglobulin with dextran through the Maillard reaction. Food Hydrocolloids, Semo, E., Kesselman, E., Danino, D., & Livney, Y. D. (2007). Casein micelle as a natural
19, 831–837. nano-capsular vehicle for neutraceuticals. Food Hydrocolloids, 21, 936–942.
Jourdain, J., Leser, M. E., Schmitt, C., & Dickinson, E. (2008a). Stability of emulsions Siew, C. K., & Williams, P. A. (2008). Role of protein and ferulic acid in the emul-
containing sodium caseinate and anionic polysaccharides. In P. A. Williams, & sification properties of sugar beet pectin. Journal of Agricultural and Food
G. O. Phillips (Eds.), Gums and stabilisers for the food industryd14 (pp. 272–279). Chemistry, 56, 4164–4171.
Cambridge, UK: Royal Society of Chemistry. Sikora, M., Badrie, N., Deisingh, A. K., & Kowalski, S. (2008). Sauces and dressings:
Jourdain, J., Leser, M. E., Schmitt, C., Michel, M., & Dickinson, E. (2008b). Stability of a review of properties and applications. Critical Reviews in Food Science and
emulsions containing sodium caseinate and dextran sulfate: relationship to Nutrition, 48, 50–77.
complexation in solution. Food Hydrocolloids, 22, 647–659. Sosa-Herrera, M. G., Berli, C. L. A., & Martinez-Padilla, L. P. (2008). Physicochemical
Kato, A., Sasaki, Y., Furuta, R., & Kobayashi, K. (1990). Functional protein/poly- and rheological properties of oil-in-water emulsions prepared with sodium
saccharide conjugate prepared by controlled dry heating of ovalbumin/dextran caseinate/gellan gum mixtures. Food Hydrocolloids, 22, 934–942.
mixtures. Agricultural and Biological Chemistry, 54, 107–112. Taherian, A. R., Fustier, P., Britten, M., & Ramaswamy, H. S. (2008). Rheology and
Kita, K., Korlos, F., & Kiosseoglou, V. (2007). Improvement, by dry-heating, of the stability of beverage emulsions in the presence and absence of weighting
emulsion-stabilizing properties of a whey protein concentrate obtained agents: a review. Food Biophysics, 3, 279–286.
through carboxymethylcellulose complexation. Food Chemistry, 104, 1153–1159. Taherian, A. R., Fustier, P., & Ramaswamy, H. S. (2007). Effects of added weighting
Leal-Calderon, L., & Schmitt, V. (2008). Solid-stabilized emulsions. Current Opinion agent and xanthan gum on stability and rheological properties of beverage
in Colloid and Interface Science, 13, 217–227. cloud emulsions formulated using modified starch. Journal of Food Process
Lopetinsky, R. J. G., Masliyah, J. H., & Xu, Z. (2006). Solids-stabilized emulsions: Engineering, 30, 204–224.
a review. In B. P. Binks, & T. S. Horozov (Eds.), Colloidal particles at liquid Taisne, L., Walstra, P., & Cabane, B. (1996). Transfer of oil between emulsion drop-
interfaces (pp. 186–224). Cambridge, UK: Cambridge University Press. lets. Journal of Colloid and Interface Science, 184, 378–390.
Matia-Merino, L., & Dickinson, E. (2004). High-sugar content acid-induced caseinate Takano, K., Hattori, M., Yoshida, T., Kanuma, S., & Takahashi, K. (2007). Porphyran as
gels and emulsion gels: influence of low-methyoxyl pectin. In P. A. Williams, & a functional modifier of a soybean protein isolate through conjugation by the
G. O. Phillips (Eds.), Gums and stabilisers for the food industryd12 (pp. 461–474). Maillard reaction. Journal of Agricultural and Food Chemistry, 55, 5796–5802.
Cambridge, UK: Royal Society of Chemistry. Tan, C.-T. (2004). Beverage emulsions. In S. E. Friberg, K. Larsson, & J. Sjöblom (Eds.),
McClements, D. J. (2005). Food emulsions (2nd ed.). Boca Raton, FL: CRC Press. Food emulsions (4th ed.). (pp. 485–524) New York: Marcel Dekker.
McNamee, B. F., O’Riordan, E. A., & O’Sullivan, M. (1998). Emulsification and Tcholakova, S., Denkov, N. D., & Lips, A. (2008). Comparison of solid particles,
encapsulation properties of gum arabic. Journal of Agricultural and Food Chem- globular proteins and surfactants as emulsifiers. Physical Chemistry Chemical
istry, 46, 4551–4555. Physics, 10, 1608–1627.
1482 E. Dickinson / Food Hydrocolloids 23 (2009) 1473–1482

Uruakpa, F. O., & Arntfield, S. D. (2005). Emulsifying characteristics of commercial Wooster, T. J., & Augustin, M. A. (2006). b-Lactoglobulin–dextran Maillard conju-
canola protein–hydrocolloid systems. Food Research International, 38, 659–672. gates: their effect on interfacial thickness and emulsion stability. Journal of
Walstra, P. (1983). Formation of emulsions. In P. Becher (Ed.), Encyclopedia of Colloid and Interface Science, 303, 564–572.
emulsion technology, Vo1. 1 (pp. 57–127). New York: Marcel Dekker. Wooster, T. J., & Augustin, M. A. (2007). Rheology of whey protein–dextran
Walstra, P., & Smulders, I. (1997). Making emulsions and foams: an overview. In conjugate films at the air/water interface. Food Hydrocolloids, 21, 1072–
E. Dickinson, & B. Bergenståhl (Eds.), Food colloids: proteins, lipids and poly- 1080.
saccharides (pp. 367–381). Cambridge, UK: Royal Society of Chemistry. Yadav, M. P., Johnston, D. B., Hotchkiss, A. T., Jr., & Hicks, K. B. (2007). Corn fibre gum:
Weinbreck, F., de Vries, R., Schrooyen, P., & de Kruif, C. G. (2003). Complex coac- a potential gum arabic replacer for beverage flavour emulsification. Food
ervation of whey proteins and gum arabic. Biomacromolecules, 4, 293–303. Hydrocolloids, 21, 1022–1030.
Williams, P. A., Sayers, C., Viebke, C., Senan, C., Mazoyer, J., & Boulenguer, P. (2005). Yadav, M. P., Parris, N., Johnston, D. B., & Hicks, K. B. (2008). Fractionation, charac-
Elucidation of the emulsification properties of sugar beet pectin. Journal of terization, and study of the emulsifying properties of corn fibre gum. Journal of
Agricultural and Food Chemistry, 53, 3592–3597. Agricultural and Food Chemistry, 56, 4181–4187.
Wollenweber, C., Makievski, A. V., Miller, R., & Daniels, R. (2000). Adsorption of Ye, A., Flanagan, J., & Singh, H. (2006). Formation of stable nanoparticles via elec-
hydroxypropyl methylcellulose at the liquid/liquid interface and the effect on trostatic complexation between sodium caseinate and gum arabic. Biopolymers,
emulsion stability. Colloids and Surfaces A, 172, 91–101. 82, 121–133.