Food Research International 69 (2015) 156–163

Contents lists available at ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Beverage emulsions: Comparison among nanoparticle stabilized

emulsion with starch and surfactant stabilized emulsions
Amol Chaudhari a, Yuanjie Pan a, Nitin Nitin a,b,⁎
a
Department of Food Science and Technology, University of California Davis, 2221 RMI South, Old Davis Road, Davis, CA 95616, United States
b
Department of Biological and Agricultural Engineering, University of California Davis, 2221 RMI South, Old Davis Road, Davis, CA 95616, United States

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

Article history: Emulsions are widely used in beverages to impart desired appearance and flavor to the products. Ring formation
Received 29 September 2014 in beverages with emulsions during thermal processing and storage is one of the key challenges. This study was
Accepted 20 December 2014 aimed at comparing the relative effectiveness of silica nanoparticle based emulsifiers with surfactant and bio-
Available online 26 December 2014
polymer based emulsifier (modified starch) in influencing physical stability of emulsions in a model juice. The
stability of emulsions was measured by characterizing changes in emulsion droplet size, zeta potential, UV–vis
Chemical compounds studied in this article:
Polyoxyethylene sorbitan monolaurate
absorbance and visual evaluation of phase separation or ring formation in both primary emulsions and beverage
(PubChem CID: 443314) emulsions as a function of storage time. The influence of thermal processing on stability of emulsions both imme-
Silica (PubChem CID: 24261) diately after processing and upon storage was evaluated. The thermal processing conditions simulated both high
Citric acid (PubChem CID: 311) temperature short time and low temperature long time pasteurization conditions. The results demonstrate that
Sodium benzoate (PubChem CID: 517055) the mean droplet diameter of primary emulsions stabilized by selected emulsifiers was stable during storage for
21 days with and without pasteurization. Based on measurements of mean droplet diameter and visible ring for-
Keywords: mation, polyoxyethylene sorbitan monolaurate (tween-20) stabilized emulsion was not stable in a model juice
Beverage emulsion
and the stability of this emulsion was further reduced with thermal processing. In contrast, starch and silica sta-
Emulsion stability
bilized emulsions in a model juice did not show significant changes in particle diameter or visible ring formation
Particle stabilized emulsions
Ring formation during storage with and without prior thermal processing, although starch stabilized emulsion did show a de-
crease in absorbance during storage. Zeta potential measurements in a model juice indicate that the surface prop-
erties of emulsions were significantly distinct from those of primary emulsion, indicating interaction of juice
components with the emulsion interface influencing the surface charge at the interface. These changes in zeta po-
tential of emulsion droplets did not correlate with reduced stability of the emulsions. Overall, the results demon-
strate that nanoparticle stabilized emulsions can improve stability of emulsion in beverages as compared to
surfactant and biopolymer stabilized emulsions and provides a comprehensive matrix to evaluate stability of
emulsions in beverages.
© 2014 Elsevier Ltd. All rights reserved.

1. Introduction emulsion. Currently, polysaccharides such as gum Arabic and modified

Emulsions are commonly used in beverage products including juice
and soft drinks to influence flavor, appearance and color of the product
(Buffo, Reineccius, & Oehlert, 2001; Given, 2009). Instability such as
phase separation of the emulsion in beverage products can influence
flavor and color profile and limit the shelf life of the product. Significant
research efforts are being made to select optimal formulations of
biopolymer emulsifiers to improve stability of emulsions in beverage
products (Piorkowski & McClements, 2013). This research has been
motivated by the understanding that selection of optimal emulsifier
influences both the diameter of emulsion droplets and stability of
⁎ Corresponding author at: Department of Biological and Agricultural Engineering,
University of California Davis, 2221 RMI South, Old Davis Road, Davis, CA 95616, United
States. Tel.: +1 39 512093082; fax: +1 39 512093785.
E-mail address: nnitin@ucdavis.edu (N. Nitin).
starch are commonly used as emulsifiers for beverage emulsions.
These polysaccharide based emulsifiers are preferred compared to
globular proteins and other small molecule emulsifiers because the
latter are more sensitive to environmental conditions (pH, ionic
strength, temperature) (Chanamai & McClements, 2002) and may also
influence the flavor profile of the product (McClements, 1998). Gum
Arabic is derived from Acacia senegal and consists of various polymer
fractions (Anderson, Howlett, & Mcnab, 1985; Goodrum, Patel,
Leykam, & Kieliszewski, 2000). The hydrophobic fractions of gum Arabic
molecules stabilize an oil droplet in water by anchoring to the oil
surface while the hydrophilic fraction of the biopolymer extends into
water and stabilize the oil droplet by steric hindrance and electrostatic
repulsion (Chanamai & McClements, 2000). Modified starch has also
been used for stabilization of beverage emulsions. Since the natural
starch backbone is hydrophilic, chemical modification such as octenyl-
succinic acid modification of natural starch is carried out to introduce

http://dx.doi.org/10.1016/j.foodres.2014.12.030
0963-9969/© 2014 Elsevier Ltd. All rights reserved.
 A. Chaudhari et al. / Food Research International 69 (2015) 156–163
hydrophobic groups for stabilization of the beverage emulsions
(Trubiano, 1995). This modification has been FDA approved for use in
food. OSA starches have also been approved as food additive in the EU
(E1450). The hydrophobic sections, as a result of modification, an-
chor the starch molecule to the oil droplet surface, while the hydrophilic
starch chains stick out into the aqueous phase and protect droplets
against aggregation through steric repulsion (Tesch, Gerhards, &
Schubert, 2002).
Despite significant application of polysaccharides in beverage
emulsions, there are some limitations of the polysaccharide based
emulsifiers used for stabilizing beverage emulsions. For stable emulsion
formation, gum Arabic has to be used at relatively high concentrations.
For example, 20% gum Arabic is required to stabilize 12.5% oil-in-water
(O/W) emulsion (Chanamai & McClements, 2002). Similar to gum
Arabic, modified starch also have low interfacial activity (Piorkowski &
McClements, 2013). Therefore, relatively higher concentration of starch
as compared to conventional surfactants such as polyoxyethylene
sorbitan monolaurate (tween-20) is required to form stable emulsions.
As a result, a significant fraction of gum Arabic or starch remains dis-
persed in the aqueous phase and is not absorbed at the oil–water inter-
face in an emulsion. These non-absorbed molecules can influence
stability of emulsions as studies have shown that non-adsorbed mole-
cules can increase the attractive forces between oil droplets due to an
osmotic effect (Chanamai & McClements, 2001). These interactions be-
tween the emulsion droplets and the unabsorbed biopolymer can result
in flocculation of the oil droplets. Another potential issue with the use of
gum Arabic is the reliability of the supply chain and natural variance in
the composition of gum Arabic (Garti, 1999).
In contrast to biopolymer based stabilizers, nanoparticles can also
stabilize oil droplets in water and form emulsions that are known as
the Pickering emulsions (Chevalier & Bolzinger, 2013; Pickering,
1907). The unique advantage of the Pickering emulsion is the near
irreversible adsorption of nanoparticles at the oil–water interface in
contrast to highly dynamic interfacial properties of the biopolymer
emulsifiers (Tcholakova, Denkov, & Lips, 2008). The restricted move-
ment of the nanoparticles compared to the biopolymer emulsifiers is
expected to improve stability of the emulsions. Particle stabilized
emulsions are commonly used in manufacturing of food products such
as mayonnaise, salad dressing, yogurt, whipped products, etc. (Leal-
Calderon, Thivilliers, & Schmitt, 2007) and are known to be efficient
stabilizers for these applications (Leal-Calderon & Schmitt, 2008). Syn-
thetic (organic and inorganic) as well as biological particles can be
used for stabilizing emulsions (Boker, He, Emrick, & Russell, 2007). In
this study, we chose silica particles for stabilization of the beverage
emulsion as silica nanoparticles can stabilize emulsion droplets over a
large range of pH conditions (Midmore, 1998) which is a prerequisite
for beverage applications. Therefore, emulsions stabilized by silica par-
ticles are suitable for beverage applications. Furthermore, significant re-
search has been done to form silica nanoparticles with precise control
on size, surface chemistry that can be used in many food and cosmetic
formulations (Gibbs, Kermasha, Alli, & Mulligan, 1999; Schutt, Klein,
Mattrey, & Riess, 2003; Tikekar, Pan, & Nitin, 2013; Zhao, Dan, Pan,
Nitin, & Tikekar, 2013).
The overall goal of this study was to compare the stability of a silica
stabilized Pickering emulsion as compared to the emulsions stabilized
using a non-ionic emulsifier such as tween-20 and a polysaccharide
emulsifier such as octenyl succinic anhydride modified starch in a
model pear juice. Tween-20 is a non-ionic surfactant while modified
starch is a relatively large MW (avg. Mw ~ 106–107 Da) biopolymer.
Many juice products including the model pear juice as used in this
study are acidic in nature. It was hypothesized that the beverage emul-
sion stabilized by inorganic silica particles provides better stability to
the emulsion in the acidic environment of the juice. In this study, stabil-
ity of the emulsion in a model beverage product was determined by in-
situ monitoring of emulsion mean droplet diameter and zeta potential
of the droplets. Since pasteurization of beverage products may be
157
desirable for extended shelf life and for elimination of pathogens, stabil-
ity of the primary emulsion and the beverage formulation with emul-
sion was also evaluated after pasteurization. The results of this study
will highlight the potential of nanoparticle based emulsifier for stabili-
zation of oil droplets in beverage products.
2. Materials and methods
2.1. Materials
Tween-20, silica (LUDOX® CL colloidal silica) and citric acid (anhy-
drous) was purchased from Sigma-Aldrich (St. Louis, MO, USA). The col-
loidal silica was supplied as 30 wt.% suspension in water. The silica
particle size in the given suspension is 20 nm and zeta potential is
+ 40 mV. Starch (CAPSUL® TA) and sodium benzoate were obtained
from the National Starch Food Innovation (Bridgewater, NJ, USA) and
Fisher Scientific (Hampton, NH, USA) respectively. CAPSUL® TA is an
octenyl-succinate anhydride starch, which is made by esterification of
starch and anhydrous octenylsuccinic acid under alkaline conditions.
The degree of modification is less than 3%. High fructose corn syrup 55
(HFCS) with 77% solids, pear juice concentrate and medium chain
triglyceride (oil) was gifted by Dr. Pepper Snapple Group (Plano, TX).
Ultrapure water (16 MΩ cm) was obtained from an in-house water fil-
tration system and used as is for all the experiments. All the materials
were used as is until and unless mentioned otherwise.
2.2. Beverage emulsion preparation
The concentrations mentioned are w/w until and unless mentioned
otherwise. Tween-20 (2%), starch (5%) and silica (5%) solutions were
prepared in ultrapure water with 0.2% citric acid and 0.3% sodium ben-
zoate. Sodium benzoate was included in the formulation in order to
avoid microbial growth in the emulsion during storage. Coarse emul-
sions were prepared by dispersing 5% oil (medium chain triglyceride)
in tween-20, starch or silica solutions using a hand-held dispers-
er (Ultra-Turrax, model T25, IKA Works, Wilmington, NC) set at
24,000 rpm for 2 min. These coarse emulsions were subsequently
homogenized by a probe sonicator (Q55, QSonica, Newtown, CT) for
30 s at amplitude of 50% to achieve uniform droplet diameter distribu-
tion. The pH of all the emulsions was measured and found out to be 4.
2.3. Model juice preparation
Model pear juice was prepared by mixing the following components
in the mentioned concentrations: ultrapure water (84%), HFCS (14%),
citric acid (0.25%), sodium benzoate (0.3%) and pear juice concentrate
(1%). The mixture was stirred at 1000 rpm for 10 min. The emulsion
(0.05%) was mixed with the model juice. The pH of the model pear
juice with emulsion remained unchanged (pH = 4).
2.4. Pasteurization conditions
Pasteurization of primary emulsions and model juice was carried out
at 2 different conditions: 70 °C for 30 min and 90 °C for 3 min. Pasteur-
ization was performed in a water bath at atmospheric pressure without
agitation of the samples.
2.5. Mean droplet diameter and zeta potential measurement
Hydrodynamic diameter and zeta potentials of oil droplets in prima-
ry emulsions as well as in model juice were measured using a particle
size analyzer (Model: Malvern Nano Series; Malvern Instruments, Inc.,
Westborough, MA). Primary emulsions were diluted 100 × in water
prior to mean droplet diameter measurement. Emulsion mean droplet
diameter in model juice was measured without further dilution. For di-
ameter measurements, the analyzer was set to following specifications:
158 A. Chaudhari et al. / Food Research International 69 (2015) 156–163
material type — oil, droplet refractive index = 1.45, dispersant type —
water, dispersant refractive index = 1.33, temperature = 25 °C. Mean
droplet diameter measurements were analyzed based on the intensity
average droplet diameter distribution. The data is represented as a
mean droplet diameter (mean ± SD) at a particular time point.
2.6. Visual observations
Surface of model juice was observed at a cross-section regularly over
the course of the experiment particularly for ring formation.
2.7. Absorbance measurement
Absorbance of model pear juice with and without emulsions and as
an effect of pasteurization was measured for 21 days at regular intervals.
The absorbance measurements were carried out to monitor the changes
in the juice turbidity as an effect of emulsion addition and pasteuriza-
tion. For the measurement, 200 μL of sample was pipetted into a clear
96-well-plate (Costar-Corning, Tewksbury, MA) and the absorbance
was measured using a Molecular Devices M2 plate-reader (Carlsbad,
CA) at 500 nm. The absorbance measurements were normalized to the
absorbance of model juice at day 0 of particular treatment of pasteuriza-
tion or no pasteurization. The normalization was carried out using fol-
lowing formula —
Normalized absorbance ¼
Absorbance
Initial absorbance at day 0 for particular pasteurization treatment
The normalization was carried out to determine the changes occur-
ring in the absorbance of the model juice as a function of emulsion, pas-
teurization condition and storage time. The data is represented as
normalized absorbance (mean ± SD) at a particular time point.
2.8. Statistical analyses
The statistical analysis for comparing datasets was carried out by
Excel (Microsoft, Seattle, WA). A single tail, two sample unequal vari-
ance t-test was used to find out differences in the absorbance mea-
surements. A significance level of 5% was chosen as a threshold for
significant differences. Statistical comparison of the normalized absor-
bance values of model juice with or without emulsions at each time
point during storage was compared. The effect of pasteurization on
absorbance of model juice with and without emulsion addition was
evaluated. It was observed that on day 0 there was no significant differ-
ence in the absorbance of model juice (controls without emulsion)
Fig. 1. (A) Mean droplet diameter, and (B) zeta potential of primary emulsions at room temperature. Data points represent mean ± SD; N = 3. The table indicates the PDI values for the
mean droplet diameter of the emulsions during storage.
samples with and without pasteurization. Therefore, the results of dif-
ferent pasteurization conditions were not statistically compared for
the samples without emulsion. Significant differences (p b 0.05) be-
tween tween-20/starch, starch/silica and silica/tween-20 were men-
tioned as *, ** and *** respectively for the particular time point. At the
end of the experiment (day 21) ‘†’ is indicated for differences compared
to the initial (day 0) absorbance measurement.
3. Results
3.1. Stability of primary emulsions
Hydrodynamic mean diameters of oil-droplets and zeta potential for
primary emulsions stabilized with tween-20, starch and silica respec-
tively at room temperature are shown in Fig. 1. Emulsion mean droplet
diameters and zeta potentials were measured for 21 days at regular in-
tervals as an indicator of emulsion stability. Mean droplet diameter
(Fig. 1A) for emulsion stabilized with tween-20 was in the range of
299 to 344 nm. For starch stabilized emulsion, the mean droplet diam-
eter varied in the range of 330 to 403 nm, while in the case of emulsion
stabilized by silica, the mean droplet diameter was in the range of 385 to
461 nm. No overall changes in the mean droplet diameters were ob-
served for individual emulsions over a period of 21 days. Zeta potential
of emulsions stabilized by tween-20, starch and silica stabilized primary
emulsions (Fig. 1B) was approximately −20, −30 and +55 mV respec-
tively and did not show variation over a period of 21 days of storage.
3.2. Stability of primary emulsion after pasteurization
Mean droplet diameters and zeta potential of the primary emulsions
after pasteurization at 70 °C are shown in Fig. 2. The tween-20 stabilized
emulsion has a smaller mean droplet diameter (Fig. 2A) on day 7 as
compared to that on day 0. A similar trend in the measurement of
mean droplet diameter of emulsion stabilized with starch was also
observed following pasteurization and storage for 7 days. The silica
stabilized emulsion did not show any variation in their mean droplet di-
ameter over 21 days of storage after pasteurization. No increase in the
PDI for the primary emulsions after the 70 °C pasteurization was ob-
served for tween-20, silica and starch stabilized emulsions during
21 days of storage. The zeta potential for all the emulsions after 70 °C
pasteurization (Fig. 2B) was stable over the course of the experiment
and the values were approximately, − 20, − 30 and + 50 mV for the
tween-20, starch, and silica stabilized emulsions respectively.
Mean droplet diameters and zeta potential of the primary emulsions
after pasteurization at 90 °C are shown in Fig. 3. For tween-20 and silica
 A. Chaudhari et al. / Food Research International 69 (2015) 156–163 159

Fig. 2. (A) Mean droplet diameter, and (B) zeta potential of primary emulsions after pasteurization at 70 °C. Data points represent mean ± SD; N = 3. The table indicates the PDI values for
the mean droplet diameter of the emulsions during storage.

stabilized emulsions the mean droplet diameters (Fig. 3A) were con-
stant for 21 days and approximately equal to 300 nm and 560 nm re-
spectively after pasteurization at 90 °C. The zeta potential for all the
emulsion droplets after 90 °C pasteurization (Fig. 3B) was stable over
the course of 21 days and the values were approximately, − 15, − 30
and +50 mV for the tween-20, starch, and silica emulsions respectively.
Overall no changes in the mean droplet diameter for tween-20 and
starch emulsion were observed upon pasteurization compared to the
primary emulsion controls (without thermal treatment as shown in
Fig. 1). The mean droplet diameter of the silica stabilized emulsion
was unchanged after pasteurization of the primary emulsions at 70 °C.
However, pasteurization of silica stabilized emulsions at 90 °C apparent-
ly increased the mean droplet diameter by approximately 100 nm when
compared to primary emulsions without pasteurization. No changes
were observed in the zeta potential of the tween-20, starch and silica
stabilized emulsion droplets with pasteurization at 70 and 90 °C.
3.3. Stability of emulsions in model pear juice
Mean droplet diameter and zeta potential of emulsions in the model
pear juice as a function of storage time are presented in Fig. 4. In the case
of tween-20 stabilized emulsion, the mean droplet diameter (Fig. 4A) in
juice was higher than that of the primary emulsion (Fig. 1A). Moreover,
during storage the mean droplet diameter of the tween-20 stabilized
emulsion increased during the first seven days of storage. No further
changes in the mean droplet diameter of emulsions stabilized with
tween-20 after seven days of storage were observed. No changes in
the mean droplet diameter of starch and silica stabilized emulsions in
the model juice were observed compared to the respective primary
emulsions for 21 days. Initial PDI of emulsion stabilized with tween-
20 was similar to that in the primary emulsion but eventually increased
to approximately 2× to 0.48 after 21 days of storage in the model juice.
The PDI for emulsions stabilized with starch or silica in the model juice
also increased as compared to the primary emulsions after initial mixing
but remained unchanged for 21 days in storage. Zeta potential for all the
emulsions in juice (Fig. 4B) changed drastically compared to the prima-
ry emulsions. For emulsions stabilized with tween-20, starch and silica,
the zeta potential changed from −20, −30 and +55 mV in the primary
emulsions to approximately −5, −5 and −15 mV respectively in the
model juice.
The impact of pasteurization conditions on stability of emulsion
droplets in the model juice is presented in Fig. 5. An increase in the
mean droplet diameter (Fig. 5A, C) of the tween-20 stabilized emulsion
in the model juice was observed after both the pasteurization treat-
ments. After 21 days of storage, the mean droplet diameter of tween-
20 emulsions decreased to approximately 400 nm. No apparent effect
of pasteurization was observed on the mean droplet diameter of starch
and silica emulsions in the model juice compared to the control primary

Fig. 3. (A) Mean droplet diameter, and (B) zeta potential of primary emulsions after pasteurization at 90 °C. Data points represent mean ± SD; N = 3. The table indicates the PDI values for
the mean droplet diameter of the emulsions during storage.
160 A. Chaudhari et al. / Food Research International 69 (2015) 156–163

Fig. 4. (A) Mean droplet diameter, and (B) zeta potential of emulsions in the model pear juice at room temperature. Data points represent mean ± SD; N = 3. The table indicates the PDI
values for the mean droplet diameter of the emulsions in the model pear juice during storage.

emulsions. Overall higher PDI was observed for both the starch and sil-
ica emulsions in juice as compared to the primary emulsions. The zeta
potential (Fig. 5B, D) of the emulsions in the juice did not change with
pasteurization in the model juice but was different from that of the pri-
mary emulsions.
3.4. Visual observations
A clear ring formation was observed at the surface of model juice
with tween-20 stabilized emulsion starting from day 3. This effect was
more pronounced in pasteurized samples. No ring formation was ob-
served for juice with starch and silica stabilized emulsions with or with-
out pasteurization treatments during 21 days of storage.
3.5. Absorbance of model juice
Changes in absorbance (Fig. 6) of juice samples at the 500 nm wave-
length were measured with and without pasteurization during a period
of 21 days. The effect of pasteurization on the absorbance of model juice
without emulsions was not observed. For the unpasteurized samples
(Fig. 6A), the results demonstrate that the model juice with 0.05% of
test emulsions had a significantly higher absorbance than the control
model juice at all-time points. Furthermore, the results indicate a de-
crease in absorbance of the tween-20 and starch stabilized emulsions
after 21 days. In contrast, the absorbance values obtained for the silica
stabilized emulsion in model juice show no significant changes as a
function of storage time. After 14 days of storage, the silica stabilized
emulsion has significantly (p b 0.05) higher absorbance as compared

Fig. 5. (A) Mean droplet diameter, and (B) zeta potential of emulsions in model pear juice after pasteurization at 70 °C. (C) Mean droplet diameter, and (D) zeta potential of primary emul-
sions in model pear juice after pasteurization at 90 °C. Data points represent mean ± SD; N = 3. The table inset in the graphs indicates the PDI values for the mean droplet diameter of the
emulsions during storage.
 A. Chaudhari et al. / Food Research International 69 (2015) 156–163
Fig. 6. Absorbance of the model pear juice with and without emulsions (A) at room temperature; (B) after 70 °C pasteurization; and (C) after 90 °C pasteurization. The data points are
normalized to the absorbance of model juice at day 0 for each condition. Data points represent mean ± SD; N = 3. On particular time points the differences in tween-20/starch,
starch/silica and silica/tween-20 are indicated by ‘*’, ‘**’ and ‘***’ respectively. ‘†’ indicates statistically significant difference when compared to the initial measurement at day 0.
to the tween-20 and starch emulsions. The results of absorbance mea-
surements in model juice with emulsion formulations after pasteuriza-
tion at 70 and 90 °C are shown in Fig. 6B and C respectively. A similar
trend in absorbance values was observed for the juice samples after
pasteurization. In the pasteurized juice samples, the absorbance values
of tween-20 and starch stabilized emulsion decreased (p b 0.05) upon
storage, while absorbance values for the silica emulsion were not influ-
enced (p N 0.05) as a function of storage time. At the end of 21 days, the
absorbance of juice with silica emulsion was higher (p b 0.05) than the
juice with starch and tween-20 stabilized emulsions, at both pasteuriza-
tion conditions.
4. Discussion
Beverages are among the fastest growing food segments. Emulsions
are used in many beverage formulations. To develop shelf-stable bever-
age formulations, it is critical to optimize the stability of emulsions in
beverage formulations during processing and storage. This study has
evaluated the role of interfacial composition of emulsion formulations
in stability of emulsions in both primary forms as well as in a model
juice system with and without thermal processing. The study evaluated
the potential of nanoparticle stabilized emulsions also known as the
Pickering emulsions for beverage formulations. The results obtained
for the Pickering emulsions were compared with emulsions stabilized
using tween-20 and modified starch respectively. The thermal process-
ing conditions were selected to model both the low temperature –
longer time pasteurization and the high temperature – short time pas-
teurization conditions.
4.1. Effect of pasteurization on stability of primary emulsions
Prior studies have evaluated stability of modified starch emul-
sions immediately after pasteurization (Chanamai & McClements,
2002; Charoen et al., 2011) or during storage (McClements, Henson,
Popplewell, Decker, & Choi, 2012; Yusoff & Murray, 2011) but there is
limited information available regarding the influence of pasteurization
on the stability of emulsions upon storage. Furthermore, there is cur-
rently no available data indicating the stability of the Pickering emul-
sions upon storage in beverages after pasteurization. To address the
gaps in knowledge, stability of emulsions in their primary form and in
beverage formulations was evaluated both with and without pasteuri-
zation during a storage period of 21 days. The results based on particle
size and zeta potential measurements demonstrate that the primary
emulsions stabilized with tween-20, starch and silica were stable during
extended storage for 21 days. Although a decrease in the mean droplet
diameter for tween-20 and starch stabilized emulsions pasteurized at
70 °C was observed, overall the mean droplet diameter was not affected
much. This observation suggests that all the emulsions selected in this
work are stable. The decrease in the mean droplet diameter may be ex-
plained by the separation of the larger oil droplets during the first week
of storage.
161
4.2. Emulsion stability in model juice
Stability of an emulsion in a model juice can be significantly different
from stability of the primary emulsions due to several factors including
differences in viscosity of the primary and the diluted emulsions in
juice, presence of the particulate matter in juice concentrate and chem-
ical and physical interactions between the emulsions and the juice
components. Currently, a qualitative and quantitative measurement of
ringing (visible ring formation) is one of the widely used approaches
in the beverage industry to evaluate stability of beverage emulsions
(Given, 2009). This method is based on the assumption that the increase
in droplet diameter will result in coalescence and thereby destabiliza-
tion of the emulsion resulting in formation of a ring structure at the sur-
face of a beverage. The key limitation of this approach is the limited
insight into changes in both the size and surface properties of emulsions
in beverage formulations. To enable measurement of these changes, sta-
bility of emulsions in this study was determined in a model pear juice by
in-situ measurement of mean droplet diameter and zeta potential at
regular intervals for 21 days. Prior to the addition of emulsions, the
particle size distribution (intensity mean) of a model pear juice was
measured. The results of these measurements show a multimodal parti-
cle size distribution with a relatively large PDI (greater than 0.5) as
shown in the Supplementary Fig. S2. After addition of 0.05% emulsion,
the mean droplet diameter distribution in the model juice was mono-
modal (Figs. S3, S4 and S5) with a relatively narrow PDI. These results
suggest that the emulsion droplets have a significant role in light scat-
tering properties of the beverage formulation so that a turbid appear-
ance can be imparted to the juice.
Based on mean droplet diameter and zeta potential measurements
in a model juice, an increase in the mean droplet diameter and zeta po-
tential was observed for tween-20 emulsion after 7 days, while both
starch and silica stabilized emulsions were relatively stable. The results
also showed significant changes in the zeta potential of both starch and
silica stabilized emulsions although the size of these emulsions was rel-
atively stable. Changes in zeta potential indicate interactions of interfa-
cial composition of starch or silica emulsions with the juice components.
These results indicated that it is critical to understand the interfacial
properties of beverage emulsions in a juice formulation and not limit
the analysis to primary emulsions. The results of zeta potential mea-
surements indicated that the starch and silica stabilized emulsions in
beverage formulations can be stable despite having near-neutral zeta
potential values contrary to common understanding (Piorkowski &
McClements, 2013). It is generally believed that emulsions with higher
net charge will have better stability (Rosoff, 1989). In this study, the
zeta potential for all the tested emulsions was stable over a period of
21 days and no effect of pasteurization was observed on zeta potential.
On the contrary, upon storage in model juice, the zeta potential of
the emulsions varied dramatically. Although dramatically varied zeta
potentials than the primary emulsions, modified starch and silica par-
ticles stabilized emulsions were stable in juice even after pasteurization
(70 °C and 90 °C). In summary, these results demonstrate that tween-20
162 A. Chaudhari et al. / Food Research International 69 (2015) 156–163
is not able to stabilize the emulsions included in juice over an extended
period of time. These results indicate the role of steric hindrance, elec-
trostatic repulsion and resistance of particles to desorb from the oil–
water interface in maintaining extended stability of silica emulsion
droplets during storage and thermal processing (Kraft et al., 2010).
These results also indicate that steric effects may be more dominant in
a model juice environment as the zeta potential values of these selected
emulsions in a model juice are similar.
4.3. Effect of pasteurization on emulsion stability in juice model
The results of this study demonstrated that the droplet size of
tween-20 stabilized emulsion increased significantly and resulted in
formation of a visible ringing in a model juice after pasteurization. Pas-
teurization of beverage with starch and silica stabilized emulsions did
not have any effect on the stability of these emulsions in the juice. It is
known from the literature that the beverage emulsions stabilized by
starch are very stable even after pasteurization (Charoen et al., 2011).
The stability is due to the thick hydrophilic interfacial layers which sta-
bilize the droplets through steric repulsion. Also, the high stability of
starch stabilized emulsion in the model juice is due to the lower mobil-
ity of the interfacial layer. Similarly, emulsions stabilized by silica parti-
cles were found to be very stable in model juice after pasteurization.
Particle stabilized emulsions are very stable due to a combination of
both steric hindrance and lack of mobility of the particles adsorbed at
the interface (Pieranski, 1980).
4.4. Stable absorbance in model juice
One of the key applications of emulsion formulation in beverages is
to provide a stable cloudy appearance to the product. An individual ab-
sorbance reading used as a measure of turbidity in this study may not
indicate the stability of emulsion in the juice. Therefore, the trend of de-
creasing absorbance combined with visual ring formation was observed
which gave a comprehensive indicator of phase separation and instabil-
ity in emulsion. The results of this study suggested that the silica emul-
sions can provide a stable turbidity as measured by the absorbance and
visual observation for ring formation of the juice over a period of
3 weeks even after pasteurization. These results showed no statistically
significant decrease in the absorbance of the model juice with starch
stabilized emulsion for a period of 3 weeks. But the overall trend indi-
cates that the absorbance value for the starch stabilized emulsion in
model juice decreased during this period. In prior research, Reiner
et al. (Reiner, Reineccius, & Peppard, 2010) have shown that turbidity
of a model beverage with emulsion stabilized by modified starch is re-
duced over a period of 4 weeks. The results of this study and Reniner
et al.'s are different from the results obtained by Dłużewska et al.
(Dłużewska, Panasiewicz, & Leszczyński, 2004), which demonstrated
that a slight increase in the turbidity was observed after 12 weeks.
These differences could be attributed to the differences in measurement
methods (absorbance measurements at different wavelengths) as well
as differences in the composition of the model juice in these studies.
Overall, the results of this study demonstrate that silica stabilized emul-
sions can provide stable turbidity in beverage emulsions and these
emulsion may perform better than the starch stabilized emulsions.
5. Conclusions
The results of this study based on droplet size and visible evaluation
of a model juice demonstrated that silica and starch stabilized emul-
sions were stable during storage. Thermal processing did not signifi-
cantly change droplet size or induce visible ring formation in starch or
silica stabilized emulsions in a model juice. In contrast, the tween-20
stabilized emulsion was not stable in the model juice, and thermal pro-
cessing of the model juice resulted in enhanced destabilization of the
emulsion. Evaluation of absorbance measurements of juice samples
during storage demonstrated that the absorbance of juice samples
with starch stabilized emulsions decreased during storage, while the ab-
sorbance of juice samples with silica stabilized emulsion was stable.
These results indicated that silica nanoparticle stabilized emulsions
were more stable in the model juice as compared to the starch and
tween-20 stabilized emulsions. Overall, the results of this study demon-
strate the potential of nanoparticle stabilized emulsions for beverage
applications.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.foodres.2014.12.030.
Acknowledgments
The authors acknowledge the funding support from the Dr. Pepper
Snapel Group (Plano, TX). The authors would like to thank Mr. Ishan
Khoche for his assistance with droplet size and zeta potential
measurements.
References
Anderson, D.M.W., Howlett, J.F., & Mcnab, C.G.A. (1985). The amino-acid composition of the
proteinaceous component of gum Arabic (Acacia senegal (L) Willd). Food Additives and
Contaminants, 2(3), 159–164. http://dx.doi.org/10.1080/02652038509373539.
Boker, A., He, J., Emrick, T., & Russell, T.P. (2007). Self-assembly of nanoparticles at inter-
faces. Soft Matter, 3(10), 1231–1248. http://dx.doi.org/10.1039/B706609k.
Buffo, R.A., Reineccius, G.A., & Oehlert, G.W. (2001). Factors affecting the emulsifying and
rheological properties of gum acacia in beverage emulsions. Food Hydrocolloids,
15(1), 53–66. http://dx.doi.org/10.1016/S0268-005x(00)00050-3.
Chanamai, R., & McClements, D.J. (2000). Impact of weighting agents and sucrose on grav-
itational separation of beverage emulsions. Journal of Agricultural and Food Chemistry,
48(11), 5561–5565. http://dx.doi.org/10.1021/Jf0002903.
Chanamai, R., & McClements, D.J. (2001). Depletion flocculation of beverage emulsions by
gum Arabic and modified starch. Journal of Food Science, 66(3), 457–463. http://dx.
doi.org/10.1111/j.1365-2621.2001.tb16129.x.
Chanamai, R., & McClements, D.J. (2002). Comparison of gum Arabic, modified starch, and
whey protein isolate as emulsifiers: Influence of pH, CaCl(2) and temperature. Journal
of Food Science, 67(1), 120–125. http://dx.doi.org/10.1111/j.1365-2621.2002.
tb11370.x.
Charoen, R., Jangchud, A., Jangchud, K., Harnsilawat, T., Naivikul, O., & McClements, D.J.
(2011). Influence of biopolymer emulsifier type on formation and stability of rice
bran oil-in-water emulsions: Whey protein, gum Arabic, and modified starch.
Journal of Food Science, 76(1), E165–E172. http://dx.doi.org/10.1111/j.1750-3841.
2010.01959.x.
Chevalier, Y., & Bolzinger, M.A. (2013). Emulsions stabilized with solid nanoparticles:
Pickering emulsions. Colloids and Surfaces A: Physicochemical and Engineering
Aspects, 439, 23–34. http://dx.doi.org/10.1016/j.colsurfa.2013.02.054.
Dłużewska, E., Panasiewicz, M., & Leszczyński, K. (2004). Effect of gum Arabic and
modified starch on stability of beverage emulsions. Food Science and Technology,
7(2), 10.
Garti, N. (1999). Hydrocolloids as emulsifying agents for oil-in-water emulsions. Journal of
Dispersion Science and Technology, 20(1–2), 327–355. http://dx.doi.org/10.1080/
01932699908943795.
Gibbs, B.F., Kermasha, S., Alli, I., & Mulligan, C.N. (1999). Encapsulation in the food indus-
try: A review. International Journal of Food Sciences and Nutrition, 50(3), 213–224.
http://dx.doi.org/10.1080/096374899101256.
Given, P.S. (2009). Encapsulation of flavors in emulsions for beverages. Current Opinion in
Colloid & Interface Science, 14(1), 43–47. http://dx.doi.org/10.1016/j.cocis.2008.01.
007.
Goodrum, L.J., Patel, A., Leykam, J.F., & Kieliszewski, M.J. (2000). Gum Arabic glycoprotein
contains glycomodules of both extensin and arabinogalactan-glycoproteins.
Phytochemistry, 54(1), 99–106. http://dx.doi.org/10.1016/S0031-9422(00)00043-1.
Kraft, D. J., de Folter, J. W., Luigjes, B., Castillo, S. I., Sacanna, S., Philipse, A. P., & Kegel, W. K.
(2010). Conditions for equilibrium solid-stabilized emulsions. The Journal of Physical
Chemistry B, 114(32), 10347–10356.
Leal-Calderon, F., & Schmitt, V. (2008). Solid-stabilized emulsions. Current Opinion in
Colloid & Interface Science, 13(4), 217–227. http://dx.doi.org/10.1016/j.cocis.2007.09.
005.
Leal-Calderon, F., Thivilliers, F., & Schmitt, V. (2007). Structured emulsions. Current
Opinion in Colloid & Interface Science, 12(4–5), 206–212. http://dx.doi.org/10.1016/j.
cocis.2007.07.003.
McClements, D.J. (1998). Appearance and flavor food emulsions: Principles, practice, and
techniques (2nd ed.). New York: CRC Press, 267–295.
McClements, D.J., Henson, L., Popplewell, L.M., Decker, E.A., & Choi, S.J. (2012). Inhibition
of Ostwald ripening in model beverage emulsions by addition of poorly water soluble
triglyceride oils. Journal of Food Science, 77(1), C33–C38. http://dx.doi.org/10.1111/j.
1750-3841.2011.02484.x.
Midmore, B.R. (1998). Preparation of a novel silica-stabilized oil/water emulsion. Colloids
and Surfaces A: Physicochemical and Engineering Aspects, 132(2–3), 257–265. http://
dx.doi.org/10.1016/S0927-7757(97)00094-0.
 A. Chaudhari et al. / Food Research International 69 (2015) 156–163
Pickering, S.U. (1907). Emulsions. Journal of the Chemical Society, 91, 2001–2021. http://
dx.doi.org/10.1039/Ct9079102001.
Pieranski, P. (1980). Two-dimensional interfacial colloidal crystals. Physical Review Letters,
45(7), 569–572. http://dx.doi.org/10.1103/PhysRevLett.45.569.
Piorkowski, D.T., & McClements, D.J. (2013). Beverage emulsions: Recent developments in
formulation, production, and applications. Food Hydrocolloids, 42, 5–41. http://dx.doi.
org/10.1016/j.foodhyd.2013.07.009.
Reiner, S.J., Reineccius, G.A., & Peppard, T.L. (2010). A comparison of the stability of bev-
erage cloud emulsions formulated with different gum acacia- and starch-based emul-
sifiers. Journal of Food Science, 75(5), E236–E246. http://dx.doi.org/10.1111/j.1750-
3841.2010.01625.x.
Rosoff, M. (1989). Specialized pharmaceutical emulsions. In H.A. Leiberman, M.M. Reiger,
& G.S. Banker (Eds.), Pharmaceutical dosage forms: Disperse systems (2nd ed.)Vol. 2.
(pp. 1–43). New York, NY: Mercel Dekker.
Schutt, E.G., Klein, D.H., Mattrey, R.M., & Riess, J.G. (2003). Injectable microbubbles as con-
trast agents for diagnostic ultrasound imaging: The key role of perfluorochemicals.
Angewandte Chemie International Edition, 42(28), 3218–3235. http://dx.doi.org/10.
1002/anie.200200550.
163
Tcholakova, S., Denkov, N.D., & Lips, A. (2008). Comparison of solid particles, globular pro-
teins and surfactants as emulsifiers. Physical Chemistry Chemical Physics, 10(12),
1608–1627. http://dx.doi.org/10.1039/B715933c.
Tesch, S., Gerhards, C., & Schubert, H. (2002). Stabilization of emulsions by OSA starches.
Journal of Food Engineering, 54(2), 167–174. http://dx.doi.org/10.1016/S0260-
8774(01)00206-0.
Tikekar, R.V., Pan, Y.J., & Nitin, N. (2013). Fate of curcumin encapsulated in silica nanopar-
ticle stabilized Pickering emulsion during storage and simulated digestion. Food
Research International, 51(1), 370–377. http://dx.doi.org/10.1016/j.foodres.2012.12.
027.
Trubiano, P.C. (1995). The role of specialty food starches in flavor encapsulation. Flavor
Technology, 610, 244–253.
Yusoff, A., & Murray, B.S. (2011). Modified starch granules as particle-stabilizers of oil-in-
water emulsions. Food Hydrocolloids, 25(1), 42–55. http://dx.doi.org/10.1016/j.
foodhyd.2010.05.004.
Zhao, Y., Dan, N., Pan, Y.J., Nitin, N., & Tikekar, R.V. (2013). Enhancing the barrier
properties of colloidosomes using silica nanoparticle aggregates. Journal of Food
Engineering, 118(4), 421–425. http://dx.doi.org/10.1016/j.jfoodeng.2013.04.030.