Journal of Food Engineering 104 (2011) 63–69

Contents lists available at ScienceDirect

Journal of Food Engineering

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

Effect of sucrose fatty acid esters on the particle characteristics and flow
properties of phytosterol nanodispersions
Wai Fun Leong a, Yaakob B. Che Man a, Oi Ming Lai b, Kamariah Long c, Mitsutoshi Nakajima d,
Chin Ping Tan a,⇑
a
Department of Food Technology, Faculty of Food Science and Technology, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
b
Department of Bioprocess Technology, Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
c
Malaysian Agricultural Research and Development Institute (MARDI), P.O. Box 12301, 50774 Kuala Lumpur, Malaysia
d
Graduate School of Life and Environmental Sciences, University of Tsukuba, Tennoudai 1-1-1, Tsukuba 305-8572, Japan

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

Article history: The effect of four different types of sucrose fatty acid esters as nonionic emulsifiers on the physicochem-
Received 19 March 2010 ical properties of water-soluble phytosterol nanodispersions was investigated. In general, the mean par-
Received in revised form 20 November 2010 ticle sizes of the prepared phytosterol nanodispersions ranged from 2.8 to 259.9 nm. The phytosterol
Accepted 28 November 2010
content in the final prepared nanodispersions ranged from 230.4 to 504.6 mg/l. All of the prepared phy-
Available online 3 December 2010
tosterol nanodispersions exhibited pseudoplastic flow behavior, with low yield stress ranging from 0.630
to 9.183 mPa and a low consistency coefficient of 0.608–88.710 mPas. Less than 1.5 ll of hexane residues
Keywords:
per liter of prepared nanodispersions was found in the prepared phytosterol nanodispersions. Transmis-
Sucrose fatty acid esters
Phytosterol nanodispersions
sion electron microscopy (TEM) demonstrated that the prepared phytosterol nanoparticles were spheri-
Mean particle size cal in shape. In general, the sucrose fatty acid esters P-1570, L-1695 and S-1570 are appropriate for use in
Flow properties the preparation of phytosterol nanoparticles with small mean particle size at monomodal distribution
% Transmittance with high clarity in appearance.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction in food, pharmaceuticals and cosmetics. For example, Calderilla-

Fajardo et al. (2006) prepared various types of emulsion systems
Emulsifiers are widely used in a variety of industries to improve that contained a dermatological drug with an SE, while Garti
emulsification and occasionally for other non-emulsification re- et al. (2000) studied the characteristics of SE microemulsions for
lated functions. The term ‘emulsifier’ refers to surface-active sub- food application. The commercial application of SEs in food indus-
stances that are capable of absorbing between two immiscible tries include, but are not limited to, confectionaries, emulsified fats
phases to facilitate the formation of an emulsion and to stabilize and oils, retrogradation of starch and low-calorie, cholesterol-free
the emulsion droplets formed (Dickinson, 1993). Small molecule fat substitutes (Plat and Linhardt, 2001). SEs are considered to be
emulsifiers are among the most commonly used emulsifiers in normal constituents of the common human diet, as SEs are hydro-
the food industry for their capability to produce very small parti- lyzed in the digestive system to normal dietary components prior
cles as a result of their fast adsorption kinetics (McClements, to absorption. (Mitsubishi Chemical Safety Institute Ltd., 1994).
2005). Being biodegradable, non-toxic and non-irritant to the skin, SEs
Sucrose fatty acid esters (SEs), commonly called sugar esters, are human and environmental friendly (Drummond et al., 2003).
are nonionic small molecule emulsifiers that contain a hydrophilic The use of SEs is therefore expected to increase greatly in the food,
sucrose group and a lipophilic fatty acid group. SEs are ester com- cosmetic and pharmaceutical industries in the near future.
pounds that are manufactured from pure sugar and vegetable oils. Phytosterols are functional lipids well known for their choles-
The chemical structures of the SEs used in this study are illustrated terol lowering effect. However, due to their water insoluble nat-
in Fig. 1. A sucrose monoester is the combination of 1 mol fatty ure, phytosterol fortification is limited to high-fat food products.
acid and 1 mol sucrose; diesters and triesters consist of 2 and Previous studies have demonstrated that consumption of esteri-
3 mol fatty acid, respectively, with 1 mol of sucrose. SEs are avail- fied phytosterols is very efficient at reducing serum cholesterol
able in a broad spectrum of hydrophilic–lipophilic balance (HLB) (Plat et al., 2000). Other studies have revealed that emulsified
values and those with values ranging from 1 to 16 are widely used phytosterols cause an increase in water solubility and have a high
efficacy for lowering serum cholesterol (Takeshita et al., 2007). It
⇑ Corresponding author. Tel.: +60 3 89468418; fax: +60 3 89423552. is believed that the development of water-soluble nanoparticulate
E-mail address: tancp@putra.upm.edu.my (C.P. Tan). phytosterols with small particle sizes may not only enable better

0260-8774/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jfoodeng.2010.11.028
64 W.F. Leong et al. / Journal of Food Engineering 104 (2011) 63–69

toluene (99.8%), potassium hydroxide, sodium chloride and sodium

sulfate were purchased from Merck (Darmstadt, Germany).
Standards for b-sitosterol (>95%), stigmasterol (95%), stigmastanol
(97.4%), campesterol (65% crystalline), brassicasterol and 5a-choles-
tane (internal standard), were supplied by Sigma–Aldrich (St Louis,
MO, USA). N-O-bis-(trimethylsilyl) trifluoroacetamine (BSTFA) was
purchased from Supelco (Bellefonte, PA, USA). Headspace 20 ml vials
and PTFE/silicon septa aluminium caps were obtained from Agilent
Technologies (Wilmington, DE, USA).
Fig. 1. Chemical structure of sucrose monoester, where R = alkyl group. Sucrose
palmitate, R = palmitic acid; sucrose laureate, R = lauric acid; sucrose oleate, 2.2. Preparation of the phytosterol nanodispersions
R = oleic acid; sucrose steareate, R = stearic acid.

Phytosterols in powder form were first dissolved in hexane

(0.5% w/v) to form the organic phase. Before mixing with the aque-
water solubility but may also enhance the dose response to
ous phase, the organic phase was kept in a 45 °C water bath for
phytosterols.
5 min to facilitate dispersion of the phytosterol. The aqueous phase
In a previous paper, we successfully optimized the production
was prepared by dissolving the SEs in deionized water. The organic
parameters for the production of water-soluble phytosterol micro-
phase was slowly added into the aqueous phase under conven-
dispersions (Leong et al., 2009) and nanodispersions, using poly-
tional homogenization (Silverson L4R, Buckinghamshire, UK) to
oxyethylene sorbitan monolaurate (Tween 20), a small molecule
produce a coarse pre-emulsion. The pre-emulsions were immedi-
emulsifier. Therefore, in the current study, we investigated the
ately subjected to high-pressure homogenization at 80 MPa (APV,
possibility of producing water-soluble phytosterol nanodispersions
Crawley, UK). Phytosterol nanodispersions were produced after
using a few selected SEs. The objective of this study was to demon-
the removal of the organic phase using a rotary evaporator (NE
strate the effects of different types of SEs on the physicochemical
1001, Eyela, Tokyo, Japan) attached to a circulating bath (Ca
properties and the flow behavior of the prepared water-soluble
1111, Eyela Cool Ace, Tokyo, Japan) under reduced pressure at
phytosterol nanodispersions.
0.25 bar. The circulating bath was set at 5 °C.

2. Materials and methods

2.3. Analysis of particle size and its distribution profile
In the current study, water-soluble phytosterol nanodispersions
Measurements of the mean particle size and the distribution
were prepared using the emulsification-evaporation technique.
profile of the phytosterol nanodispersion were conducted after
This technique is commonly used to produce drug particles and
the organic phase was removed. Measurements were made using
it has recently been applied in the production of functional lipid
a dynamic light-scattering particle size analyzer (Zetasizer Nano
nanodispersions (Cheong et al., 2008; Yuan et al., 2008; Chu
ZS; Malvern Instruments Ltd., Worcestershire, UK). The measure-
et al., 2007; Tan and Nakajima, 2005a). The processing and formu-
ment temperature was set at 25 °C. Measurements were taken by
lation parameters, which included conventional mixing time and
placing the nanodispersion-containing cuvette directly into the
mixing speed, high-pressure homogenization (Leong et al., 2009;
module.
Floury et al., 2000), the concentrations of the emulsifiers and the
loaded active compound, and the organic to aqueous phase ratio
(Cheong et al., 2008; Tan and Nakajima, 2005b) are known to have 2.4. Gas Chromatography analysis of phytosterols
effects on the particle size and its distribution profile. To study the
effect of SEs on the production of phytosterol nanodispersions, the 2.4.1. Sample preparation
mixing time and mixing speed were fixed to 5 min and 5000 rpm. A 300-ll sample was placed into a 250 ml flat-bottom flask.
The high-pressure homogenization step was set at 80 Mpa for one After adding 100 ll 5a-cholestane (1 g/ml) and 5 ml 1 M ethanolic
cycles. The organic to aqueous ratio was fixed at 1:9 and the con- KOH, the sample was saponificated at 90 °C for 60 min. The reacted
centration of phytosterols in the organic phase was 0.5% w/v. mixture was transferred into a 100 ml separating funnel for further
extraction. First, 10 ml hexane was added to extract the unsapon-
2.1. Materials ifiable fraction. The organic phase was then washed three times
with 5 ml distilled water to remove the soap of resins and fatty
VegapureÒ FTE (minimum 99% phytosterols) was a gift from acids. The organic phase was collected and sodium sulfate was
Cognis (Duesseldorf-Holthausen, Germany). All nonionic emulsifi- added to absorb moisture. The remaining amount of solution con-
ers of SEs were kindly contributed by Mitsubishi-Kagaku Food Corp taining phytosterol was transferred to 15 ml tubes and dried under
(Tokyo, Japan). The emulsifiers used were sucrose stearate a nitrogen stream. The dry residue was derivatized with 50 ll of
(S-1570), sucrose palmitate (P-1570), sucrose laureate (L-1695) BSTFA and incubated in a water bath at 60 °C for 60 min. A total
and sucrose oleate (OWA-1570). The HLB and percentages of of 1 ll of the mixture was analyzed by gas chromatography.
monoester concentration of the studied SEs are listed in Table 1.
HPLC grade hexane (purity P 98.8%), ethanol, dichloromethane, 2.4.2. Preparation of standards and internal standards
Standards of b-sitosterol, b-sitostanol, stigmasterol, campester-
ol and brassicasterol were prepared in ethanol at 1 mg/ml. The
Table 1
internal standard, 5a-cholestane was prepared in dichloromethane
Types of sucrose fatty acid esters (SEs) used.
at 1 mg/ml. All standards and the internal standards were analyzed
SEs Abbreviation HLB Monoester (%) immediately upon dilution. A 100 ll standard and a 100 ll internal
Sucrose laureate L-1695 16 80 standard were placed into 15 ml tubes and dried under a nitrogen
Sucrose palmitate P-1570 15 70 stream. The dry residue was derivatized with 50 ll BSTFA and
Sucrose stearate S-1570 15 70
incubated in a water bath at 60 °C for 60 min. A total of 1 ll of
Sucrose oleate OWA-1570 15 70
the mixture was analyzed by gas chromatography.
 W.F. Leong et al. / Journal of Food Engineering 104 (2011) 63–69 65

2.4.3. Gas chromatography analysis of phytosterols were ramped for 60 s. All samples were allowed to rest for
The phytosterols were analyzed using the Agilent Technologies 10 min after loading to allow for temperature equilibration and
network GC system, series 6890 N (Wilmington, DE, USA) with for the induced stress to relax. Each measurement was repeated
flame ionization detection (FID) controlled by the Agilent Chemsta- twice. Data were analyzed using the Rheowin Data Manager soft-
tion software. Separation was conducted on a MDN-5 fused silica ware, version 4.00.
capillary column of 30 m length  0.32 mm i.d  0.25 lm film
thickness (Supelco, Bellefonte, PA, USA). Splitless injection was 2.7. Transmittance measurement
used. The oven temperature program was initially set at 250 °C
and was increased at 2 °C/min to 300 °C. The detector and injection The % transmittances of the phytosterol nanodispersions were
port temperatures were set at 300 °C. measured using a UV–visible spectrophotometer (Spectronic
Genesys™ 10, GENEQ Inc., Montreal, Canada). The measurements
2.5. Determination of hexane residue were made using a standard single-beam arrangement. The
nanodispersions were place in a cuvette with a 1 cm path length.
2.5.1. Headspace solid-phase microextraction (HS-SPME) analysis Deionized water was used as blank. The transmittance spectra were
HS-SPME coupled with GC was previously used to analyze sol- obtained over the wavelength range 380–750 nm.
vent residues in edible oils (Michulec and Wardencki, 2005). A
Supelco 75 lm Carboxen polydimethylsiloxane (CAR-PDMS)-
2.8. Transmission electron microscopy (TEM)
coated SPME fiber and a manual fiber holder were used for the
SPME headspace analysis. The CAR-PDMS fiber was first condi-
The morphology and the structure of the phytosterol nanopar-
tioned in the GC injection port at 250 °C for 2 h. A total of 1.5 g
ticles were observed using a Hitachi H-7100 TEM (Tokyo, Japan).
NaCl was added to a 20 ml headspace vial containing 4 ml of phy-
The copper microscope grid coated with carbon film was first
tosterol nanodispersion. Toluene (1 ll) was used as an internal
placed into a drop of phytosterol nanodispersion for 3 min and
standard. The vial was immersed in a water bath for heating and
dried in open air. The dried sample was then negatively stained
was stirred for 15 min at 80 °C. After 15 min, the fiber was exposed
by a 2% phosphotungstic acid (PTA) solution for 2 min. The micro-
for 15 min at 80 °C for absorption, and it was then immediately in-
scope grid was then left to dry for 2 h. The dried and stained micro-
jected into the injection port of the GC for 10 min. Each analysis
scope grids containing samples were observed by TEM with an
was carried out in duplicate. To determine the relative response
acceleration voltage of 100 kV.
factor (RRF), 5 ml of the phytosterol nanodispersion was replaced
by deionized water containing hexane (1 ll).
2.9. Statistical analysis
2.5.2. Gas Chromatography analysis of hexane residue
The analysis was performed with an Agilent Technologies net- Data were analyzed using a one-way analysis of variance in
work GC system, series 6890 N (Wilmington, DE) with flame ioni- SPSS (package release 12.0). All experiments were carried out in
zation detection, controlled by a MDN-1 fused silica capillary duplicate. The significance level was set at 0.05. The significance
column of 30 m length  0.32 mm i.d  0.25 lm film thickness between means was further determined by Duncan’s multiple
(Supelco, Bellefonte, PA, USA). A 0.75 mm i.d. liner was used to range tests.
minimize the effect of peak broadening. Splitless injection was
used. Helium was used as the carrier gas at a flow rate of 1.1 ml/ 3. Results and discussion
min. The injector temperature was set at 250 °C. The oven temper-
ature program was initially set at 40 °C for 3 min, and it was then 3.1. Particle size and distribution
increased at 4 °C/min to 60 °C and then 20 °C/min to 100 °C. The
detector temperature was set at 250 °C. The hydrogen and air flows The optimum HLB value of the emulsifiers required to fully en-
were set at 45 ml/min and 400 ml/min, respectively. trap the compound of interest often depends on the hydrophilicity
of the compound. The ideal HLB value for oil-in-water dispersion is
2.6. Viscosity measurement between 9 and 18. At low HLB value (between 7 and 9), the
emulsifier may behave as a wetting agent instead of an effective
Viscosity was measured using of a controlled-stress rheometer emulsifier, which causes the entrapment to be less efficient
(RheoStress 6000, Haake GmbH, Karlsruhe, Germany). The viscosi- (McClements, 2005). The HLB value of the SE reduces as the degree
ties of the sample were evaluated at 25 °C with a measuring cup of esterification of the sucrose increases and/or the chain length of
Z43 and rotor Z38. The sample size was 30.8 ml. Temperature the fatty acid increases. To produce oil-in-water nanodispersions,
was maintained at 25 °C by a Haake temperature controller SEs with high HLB values or highly hydrophilic properties has been
(DC30) with an accuracy of ± 0.1 °C. Sheer rates from 0 to 10 s1 suggested (Garti et al., 1999).

Table 2
Particle size and its distribution of phytosterol nanodispersions prepared by different types of SEsz.

% SEs SEs
(w/v)
P-1570 L-1695 OWA-1570 S-1570
Mean particle Distribution Mean particle Distribution Mean particle Distribution Mean particle Distribution
diameter (nm) profile diameter (nm) profile diameter (nm) profile diameter (nm) profile
0.1 134.9 ± 1.5ab Bimodal 66.6 ± 0.3ac Bimodal 245.2 ± 5.40aa Bimodal 259.9 ± 6.0aa Polymodal
0.5 6.5 ± 0.1bb,c Monomodal 3.9 ± 0.6bc Monomodal 108.7 ± 1.0ca Polymodal 8.5 ± 1.1bb Bimodal
1.0 3.6 ± 0.5bb Monomodal 3.5 ± 0.1bb Monomodal 100.5 ± 1.6ca Polymodal 5.2 ± 0.6bb Bimodal
5.0 4.2 ± 0.7bb Monomodal 4.1 ± 0.2bb Monomodal 204.5 ± 2.5ba Polymodal 2.8 ± 0.2bb Monomodal
z
Each value in the table represents the mean ± standard deviation of six measurements of two replicates. Means within the same column with different superscript are
significantly different (p < 0.05). Means within the same row with different subscript are significantly different (p < 0.05). Abbreviations: see Table 1.
66 W.F. Leong et al. / Journal of Food Engineering 104 (2011) 63–69

Table 2 shows the mean particle size and distribution profile 3.2. Retention of phytosterol
of the prepared phytosterol nanodispersions. The particle sizes
ranged from 2.8 to 259.9 nm. Generally, the phytosterol nanodi- In general, the phytosterol content in the prepared nanodisper-
spersions prepared using 0.5%, 1% or 5% L-1695 or P-1570 and sions ranged from 230.4 to 504.6 mg/l (Table 3). Phytosterol con-
5% S-1570 were the most desirable for their small mean particle centrations did not differ significantly among the phytosterol
size and monomodal distribution profile (p < 0.05). OWA-1570 nanodispersions prepared with P-1570, L-1695, or S-1570 at 0.1–
demonstrated inferior entrapment efficiency by producing mean 5% w/v SE and 1% or 5% w/v of OWA-1570 (p > 0.05). Although
particle sizes greater than 100 nm at non-monomodal distribu- 0.1% w/v of P-1570, L-1695 and S-1570 produced larger particle
tion for all experimental OWA-1570 concentrations. The ability (more than 50 nm), the phytosterol retention were compatible to
of L-1695 to produce a small mean particle size (<100 nm) at 0.5, 1, and 5% w/v of SE. This observation indicated the good emul-
all experimental concentrations was the result of its particle sifying properties of the P-1570, L-1695 and S-1570. The entrap-
entrapment and stabilization efficacies. This can be explained ment efficiency of OWA-1570 was not satisfactory at 0.1% w/v or
by the high HLB value of L-1695, which is attributed to the high 0.5% w/v, where the amount of phytosterol entrapped and dis-
amount of monolaurate (80%) and lower amounts of di-, tri-, and persed in the continuous phase was significantly lower (p < 0.5).
polylaurates (20%) as well as the short chain length of lauric The dispersions prepared with 0.1% and 0.5% w/v OWA-1570 had
acid. Although P-1570, S-1570 and OWA-1570 share the same significant flaking out of the phytosterols during the preparation
HLB value of 15 and the same amount of monoesters, P-1570 process. As mentioned previously, the low entrapment efficiency
appeared to entrap the phytosterol nanoparticles more effi- of the OWA-1570 was due to its formulation. The concentration
ciently and produced significantly smaller mean particle sizes of sucrose oleate in OWA-1570 was 40% of the total weight with
(p < 0.05) at a monomodal distribution. This is because the pal- 70% sucrose monooleate. In other words, at 0.1% and 0.5% w/v of
mitic acid found in the P-1570 had a shorter chain length than OWA-1570, the amount of sucrose monooleate is not sufficient to
the stearic and oleic acids. OWA-1570 was unable to produce fully entrap and retain the phytosterols in the dispersions.
small mean particle sizes at a monomodal distribution. This The un-entrapped phytosterols were not dispersed in the aqueous
could be explained by the formulation of OWA-1570 itself. The phase and instead hovered on the surface of the prepared dispersion.
OWA-1570 formulation contained 40% sucrose oleate with 56% The dispersions prepared with 1% w/v and 0.5% w/v OWA-1570
water and 4% ethanol. The low entrapment efficiency at the therefore contained lower concentrations of phytosterols.
low concentrations (0.1% w/v, 0.5% w/v and 1% w/v) of OWA-
1570 was therefore expected due to the low concentration of su- 3.3. Hexane residue
crose oleate. The ethanol present in the OWA-1570 formulation
acted as a cosurfactant to increase the solubility of sucrose ole- Solvent residues are often the major concern in nanodispersion
ate in the water phase. It has previously been reported that the preparation by the emulsification-evaporation, solvent displace-
cosurfactant ethanol was able to increase the HLB value of the ment or emulsification-diffusion techniques. The amounts of
system by raising the relative affinity for water of the nonionic organic solvent residues directly affect the quality of the final prod-
SEs (Garti et al., 1999). However, in the current study, the cosur- ucts. It is undeniable that complete removal of the organic phase
factant did not improve the entrapment efficiency. A large mean from the final product was rather difficult and sometimes imprac-
particle size was seen at the non-monomodal distribution using tical or impossible. Hexane is commonly used as a solvent in food
5% w/v of OWA-1570. Small mean particle sizes in the nanome- processing and the extraction of vegetable oil. The detectable hex-
ter-range are preferable due to their stability against creaming, ane residue in common soy protein, soy oil and soy meal have been
sedimentation and flocculation (Solans et al., 2005; Tadros shown to range from 20 to 1000 ppm (Woerfel and Erickson, 1995).
et al., 2004). A narrow monomodal distribution profile was
required to reduce the instability due to Ostwald ripening Table 3
(McClements, 2005; Tadros et al., 2004; Taylor, 1995). Fig. 2 Phytosterols concentration of the nanodispersions prepared using different types of
sucrose fatty acid estersz.
shows the particle distribution profile of the phytosterol nanod-
ispersions prepared using 1% SEs. % SEs (w/v) Phytosterol concentration (mg/l)
P-1570 L-1695 OWA-1570 S-1570
0.1 488.0 ± 3.2a 488.2 ± 1.2a 230.4 ± 4.3b 495.0 ± 4.2a
0.5 489.2 ± 5.3a 485.0 ± 3.2a 254.4 ± 11.3b 487.0 ± 1.8a
30
1.0 492.2 ± 1.8a 503.9 ± 8.1a 489.1 ± 5.7a 500.5 ± 3.3a
5.0 490.9 ± 9.2a 492.5 ± 3.3a 504.6 ± 5.4a 502.0 ± 6.2a
25 z
Each value in the table represents the mean ± standard deviation of four mea-
surements of two replicates. Means within the same column with different sub-
script are significantly different (p < 0.05). Abbreviations: see Table 1.
20
Voulume (%)

P-11570-1%
L-1695-1%
O-1579-1% Table 4
15
Hexane residue of the phytosterol nanodispersions prepared using different types of
S-1570-1%
sucrose fatty acid estersz.
10 % SEs (w/ Hexane residue (ll/l)
v)
P-1570 L-1695 OWA-1570 S-1570
5
0.1 0.010 ± 0.002b 0.012 ± 0.003a n.d** 0.008 ± 0.002b
0.5 n.d** n.d** 0.006 ± 0.002b 0.006 ± 0.000b
0 1.0 0.005 ± 0.001b 0.007 ± 0.000a 0.006 ± 0.001b 0.006 ± 0.001b
1 10 100 1000 5.0 1.430 ± 0.004a 0.011 ± 0.002a 0.368 ± 0.002a 1.502 ± 0.003a
z
Particle diameter (nm) Each value in the table represents the mean ± standard deviation of four mea-
surements of two replicates. Means within the same column with different sub-
Fig. 2. Particle size distribution profile of phytosterol nanodispersions prepared by script are significantly different (p < 0.05). Abbreviations: see Table 1.
**
1% w/v of SEs. Not detected.
 W.F. Leong et al. / Journal of Food Engineering 104 (2011) 63–69 67

Table 4 shows the amount hexane residue in the phytosterol nan-

odispersions. Hexane residue was recorded at 0.012 ll or less per
liter of prepared phytosterol nanodispersions with 1.0% w/v or less
of SE. Hexane residues were significantly higher (p < 0.05) when
the nanodispersions were prepared with 5% w/v of all types of
SE, except for 5% w/v L-1695. Heating at reduced pressure has facil-
itated the removal of hexane. However, the hydrophobicity of the
hexane has caused the attraction of hexane molecule to the hydro-
phobic core of the sucrose ester micelle and hydrophobic tail of
phytosterol molecule. Although evaporation of hexane residual at
higher temperature could be more efficient, the increase in tem-
perature can reduce the stability of the prepared dispersions.

3.4. Flow property

Table 5 shows the yield stress (s0), consistency coefficient (K), Fig. 3. Flow Curves of phytosterol nanodispersions (filled symbols) and aqueous
and flow index (n) obtained from the flow curve models of the pre- phase containing SEs only (open symbols) at 1.0% w/v of different types of SEs.
pared phytosterol nanodispersions, which were calculated using
the Herschel–Bulkley model.
ispersions had significantly (p < 0.5) higher yield stress values than
The Herschel–Bulkley model equation is as follows:
L-1695 and P-1570. This is because S-1570 has a higher molecular
s þ s0 þ KðcÞn ð1Þ mass. The consistency coefficient (K) measures the viscous nature
of the nanodispersions. The K values were significantly higher at
where s = shear stress and c = shear rate. 5% w/v SEs for all types of SEs (p < 0.5). The OWA-1570-stabilized
The coefficients of determination (R2 P 0.95 for all samples) phytosterol nanodispersions exhibited significantly higher
indicate the fitness of the equation. The recorded yield stress ran- K values at all OWA-1570 concentrations (p < 0.05). Generally,
ged from 0.63 to 9.18 mPa. The yield stress measured the mini- L-1695-stabilized phytosterol nanodispersions had significantly
mum stress that must be exceeded before the sample was able lower yield stress values and consistency coefficients at 1.0% w/v
to flow. The densities of the structure of the emulsion network and 5.0% w/v L-1695 (p < 0.5). The flow index is an indicator of
and the molecular weight of the ingredients used have been the deviation from a Newtonian fluid. Newtonian fluid has a corre-
shown to be important factors that determine the yield stress sponding n value of 1, while n < 1 corresponds to shear-thinning
(Tadros, 2004). Low yield stress implies that the prepared phytos- behavior and n > 1 corresponds to shear thickening behavior. The
terol nanodispersions were able to flow easily when they were results obtained for all SEs stabilized phytosterol nanodispersions
poured. The low yield stress was expected due to the low volume were consistent with pseudoplastic flow behavior (n < 1). The n
fraction of the dispersed phases, the small molecular mass of the values ranged from 0.454 to 0.919.
SEs, and the low SEs concentrations used (0.1–5% w/v). Generally, The mean particle size and its distribution profile have been re-
yield stress was significantly higher at 5% SEs for all types of SEs ported to have a significant effect on the rheological properties of
used (p < 0.5). This is probably related to the increase in SEs net- the emulsion system (Nor Hayati et al., 2007; Sato and Cunha,
work structure present in the prepared nanodispersions. The yield 2009). However, this effect was barely noticeable in the current
stress was significantly lower at 1.0% w/v L-1695 and P-1570 due study due to the low dispersion phase volume fraction. Fig. 3
to the small mean particle sizes (p < 0.5). Though S-1570 also shows the rheograms for a flow curve of viscosity vs. shear rate
exhibited a small mean particle size at 1.0% w/v S-1570, its nanod- for the SEs-stabilized phytosterol nanodispersions (filled symbols)
and the aqueous phase containing solely SE (open symbols). The
Table 5 rheogram shows that the viscosity reduces as the shear rate in-
Modeling of the flow curve between 1 and 10 s1 of shear rate of the phytosterol
creases, indicating the pseudoplastic behavior of the SEs-stabilized
nanodispersions prepared by different types of sucrose fatty acid esters using
Herschel–Bulkley model: s = s0 + K (c)n.z phytosterol nanodispersions and the SEs-containing aqueous
phase. The viscosity of the phytosterol-containing nanodispersions
Concentration (% w/v) SEs
was higher than the corresponding SE-containing aqueous phase at
P-1570 L-1695 OWA-1570 S-1570 a shear rate of 10/s. At a low shear rate, the viscosity of the phytos-
Yield stress, s0 (mPa) terol nanodispersions was also higher than the continuous phase,
0.1 3.147 ba 2.600bb 2.331bb 1.778dc except for OWA-1570. This indicates that the viscosity of
0.5 3.092 ba 1.928bb 2.149bb 3.098ca the SEs-stabilized phytosterol nanodispersions was directly
1.0 1.403cc 0.630cd 2.183bb 4.141ba
5.0 6.631ab 3.003ad 4.746ac 9.183aa
proportional to its corresponding continuous phase and that the
phytosterol nanoparticles contributed to the increase in the nanod-
Consistency coefficient, K (mPas)
0.1 0.608dc 1.379cb 6.339da 1.496db
ispersion viscosity. The low yield stress and low consistency of the
0.5 3.075cd 6.946bc 11.910ca 8.428cb prepared dispersions are suitable for use in low viscosity food sys-
1.0 8.588bb 7.966ac 13.240ba 13.570ba tem such as beverages. However phytosterol nanodispersions with
5.0 63.270ab 7.931ad 88.710aa 25.830ac increase viscosity value may changes the mouth feel of the final
Flow index, n beverages.
0.1 0.919aa 0.640bc 0.670cc 0.738bb
0.5 0.454dd 0.651bc 0.810aa 0.760bb
1.0 0.518cc 0.675a,bb 0.720ba 0.746ba
3.5. Transmittance spectrum
5.0 0.865ba 0.702ab 0.688b,cb 0.838aa
z
The transmittance spectra of the prepared phytosterol nanodi-
Data are reported as means from two measurements of two replicates. Means
within the same column with different superscript are significantly different
spersions were obtained in the visible spectrum of human eyes,
(p < 0.05). Means within the same row with different subscript are significantly which ranges from 380 to 750 nm (Penner, 2003a). The interaction
different (p < 0.05). Abbreviations: see Table 1. between the nanodispersions and the electromagnetic radiation in
68 W.F. Leong et al. / Journal of Food Engineering 104 (2011) 63–69

the visible spectrum determines how the nanodispersion appears and OWA-1570 was the most opaque. Fig. 5 agrees with the %T
to the human eye. Nanodispersions are visually transparent or spectrum measurement for clarity comparison. The degree of
semi-transparent due to their small particle sizes (Solans et al., transparency of the prepared nanodispersions is strongly related
2005). The degree of transparency of the nanodispersions is depen- to the efficiency of the particle in scattering light. According to
dent on the particle size in the dispersion as well as the concentra- the Rayleigh theory (Hutchings, 1999), the intensity of the scat-
tion of the particles and the nature of its ingredients. The % tered light (Is) can be described as follows:
transmittance (%T) is the measure of the amount of light, at a spe-
cific wavelength, that passes through the sample without being ab- 8p4 r 6 ðm4  2m2 þ 1Þð1 þ cos2hÞ
2
ð3Þ
sorbed. The %T can be described by the following equation (Penner, d k4 ðm4 þ 4m4 þ 4Þ
2003b):
where r = radius of particle, k = wavelength of light, d = distance
%T ¼ ðIT =Io Þ  100 ð2Þ between particle and detector, m = relative refractive index,
h = scattering angle.
where IT = intensity of the light passing through the sample, For a fixed particle volume fraction, the low ratio of particle size
Io = intensity of the incident light. relative to the wavelength of the light, at r/k < 0.1, proposed low
In the current study, the amount of phytosterols dispersed was scattering efficiency. The nanodispersions therefore appeared to
fixed at 0.5% w/v, while the types of the SEs used were varied. Fig. 4 be highly transparent. The scattering intensity was higher as the
shows the %T spectra of the phytosterol nanodispersions prepared particle size increased. Hence, OWA-1570 nanodispersions that
with 1% SE, where the mean particle size ranged from 3.6 to contained larger particles appeared to be opaque. Generally, the
110.5 nm. In this study, the %T spectra were used to compare the %T was lower at the low wavelength measurements. This is be-
optical clarity of the prepared nanodispersions and not to quantify cause the intensity of the scattered light was inversely proportion-
the turbidity or the color. The %T spectrum of L-1695 was the high- ate to the power of four of the wavelength, as Eq. (3).
est at all measured wavelengths and was followed by P-1570 and
S-1570; OWA-1570 had the lowest %T spectrum. In other words,
nanodispersions prepared with 1% L-1695 had the highest clarity, 3.6. TEM analysis

A representative TEM picture of the phytosterol nanodisper-

sions prepared with L-1695 is shown in Fig. 6. The result of the
microscopic analysis shows that the particles are spherical in
shape. The analysis was comparable to the particle size results.

4. Conclusion

In summary, it was possible to produce water-soluble phytos-

terol nanodispersions with particle sizes less than 100 nm at a
monomodal distribution using SEs L-1695, P-1570 and S-1570.
The influence of the SEs on the properties of the phytosterol nan-
odispersions was studied. The findings showed that the amount
of phytosterols retained in the prepared phytosterol nanodisper-
sions were >485 mg/l, except at a low concentration of
OWA-1570. There was <1.5 ll hexane residue per liter of prepared
nanodispersion. The flow properties of the phytosterol nanodisper-
sions were influenced by the nanoparticles it contained and the
amount and types of SEs used. Overall, the low yield stress and
Fig. 4. Transmittance spectrum of phytosterol nanodispersions prepared by 1.0% w/ low consistency coefficient indicated that the prepared phytosterol
v of different types of SEs. nanodispersions resembled water with low viscosity and could

Fig. 5. Phytosterol nanodispersions prepared by 0.5% w/v of phytosterols and 1.0% w/v of different types of SEs.
 W.F. Leong et al. / Journal of Food Engineering 104 (2011) 63–69 69

Drummond, C.J., Boyd, B., Baker, I.J.A., 2003. Sugar fatty acid esters. In: Holmberg, K.
(Ed.), Novel surfactants: preparation, applications, and biodegradability. CRC,
New York, USA, pp. 95–128.
Floury, J., Desrumaux, A., Lardières, J., 2000. Effect of high-pressure homogenization
on droplet size distributions and rheological properties of model oil-in-water
emulsions. Innovative Food Science and Emerging Technologies 1, 127–134.
Garti, N., Clement, V., Fanun, M., Leser, M., 2000. Some characteristics of sugar ester
nonionic microemulsions in view of possible food applications. Journal of
Agricultural and Food Chemistry 48, 3945–3956.
Garti, N., Clement, V., Leser, M., Aserin, A., Fanun, M., 1999. Sucrose ester
microemulsions. Journal of Molecular Liquids 80, 253–296.
Hutchings, J.B., 1999. Food color and appearance. Aspen Publisher, Boca Raton.
Leong, W.F., Che Man, Y., Lai, O.M., Long, K., Misran, M., Tan, C.P., 2009. Optimization
of processing parameters for the preparation of phytosterol microemulsions by
the solvent displacement method. Journal of Agricultural and Food Chemistry
57, 8426–8433.
McClements, D.J., 2005. Food emulsion: Principles Practices and Techniques. CRC
Press, London.
Michulec, M., Wardencki, W., 2005. Development of headspace solid-phase
microextraction–gas chromatography method for the determination of
solvent residues in edible oils and pharmaceuticals. Journal of
Chromatography A 1071, 119–124.
Mitsubishi Chemical Safety Institute Ltd., 1994. Clinical and Pharmacokinetic
Studies of Sucrose Esters of Fatty Acids in Human. Yokohama, Japan.
Nor Hayati, I., Che Man, Y., Tan, C.P., Nor Aini, I., 2007. Stability and rheology of
concentrated o/w emulsions based on soybean oil/palm kernel olein blends.
Food Research International 40, 1051–1061.
Penner, M.H., 2003a. Basic principles of spectroscopy. In: Nielson, S.S. (Ed.), Food
Fig. 6. Transmission electron microscopy image of a phytosterol nanodispersions Analysis. Plenum Pub Corp, New York, pp. 359–370.
Penner, M.H., 2003b. Ultraviolet, visible and flourescence spectroscopy. In: Nielson,
prepared with 0.5% w/v phytosterols and 1.0% w/v L-1695.
S.S. (Ed.), Food analysis. Plenum Pub Corp, New York, pp. 371–386.
Plat, J., Van Onselen, E.M.N., Van Heugten, M.M.A., Mensink, R.P., 2000. Effects on
serum lipids, lipoproteins and fat-soluble antioxidant concentrations of
flow effortlessly. The phytosterol nanodispersions prepared by SE consumption frequency of margarines and shortenings enriched with plant
at particle sizes <100 nm had higher clarity. stanol esters. European Journal of Clinical Nutrition 54, 671–677.
Plat, T., Linhardt, R.J., 2001. Syntheses and applications of sucrose-based esters.
Journal of Surfactants and Detergents 4, 415–421.
Acknowledgments Sato, A.C.K., Cunha, R.L., 2009. Effect of particle size on rheological properties of
jaboticaba pulp. Journal of Food Engineering 91, 566–570.
Solans, C., Izquierdo, P., Nolla, J., Azemar, N., Garcia-Celma, H.J., 2005. Nano-
Financial support of this work by the Ministry of Science, emulsions. Current Opinion in Colloid & Interface Science 10, 102–110.
Technology and Innovation of Malaysia through Science Fund Tadros, T., 2004. Application of rheology for assessment and prediction of the long-
(05-01-04-SF0384) and the Ministry of Higher Education through term physical stability of emulsions. Advances in Colloid and Interface Science
108, 227–258.
Fundamental Research Grant Scheme (02-11-08-0619FR) is grate- Tadros, T., Izquierdo, P., Esquena, J., Solans, C., 2004. Formation and stability of
fully acknowledged. The first author thanks the Graduate School of nano-emulsions. Advances in Colloid and Interface Science 108, 303–318.
Universiti Putra Malaysia for the Graduate Research Fellowship. Takeshita, M., Saito, S., Katsuragi, Y., Yasunaga, K., Matsuo, N., Tokimitsu, I.,
Yasukawa, T., Nakamura, H., 2007. Combination of plant sterols and
diacylglycerol oil lowers serum cholesterol and lipoprotein (a) concentrations
References in postmenopausal women with mild-to-moderate hypercholesterolemia.
Clinical Nutrition 2, 4–11.
Calderilla-Fajardo, S.B., Cazares-Delgadillo, J., Villalobos-Garcia, R., Quintanar Tan, C.P., Nakajima, M., 2005a. Effect of polyglycerol esters of fatty acids on
Guerrero, D., Ganem-Quintanar, A., Robles, R., 2006. Influence of sucrose physicochemical properties and stability of b-carotene nanodispersions
esters on the in vivo percutaneous penetration of octyl methoxycinnamate prepared by emulsification/evaporation method. Journal of the Science of
formulated in nanocapsules, nanoemulsion, and emulsion. Drug Development Food and Agriculture 85, 121–126.
and Industrial Pharmacy 32, 107–113. Tan, C.P., Nakajima, M., 2005b. b-Carotene nanodispersions: preparation,
Cheong, J.N., Tan, C.P., Che Man, Y., Misran, M., 2008. Tocopherol nanodispersions: characterization and stability evaluation. Food Chemistry 92, 661–671.
Preparation, characterization and stability evaluation. Journal of Food Taylor, P., 1995. Ostwald ripening in emulsions. Colloids and Surfaces A:
Engineering 89, 204–209. Physicochemical and Engineering Aspects 99, 175–185.
Chu, B.S., Ichikawa, S., Kanafusa, S., Nakajima, M., 2007. Preparation of protein- Woerfel, J.B., Erickson, D.R., 1995. Practical Hand-Book of Soybean Processing and
stabilized b-Carotene nanodispersions by emulsification–evaporation method. Utilization. AOCS Press, USA.
Journal of the American Oil Chemists’ Society 84, 1053–1062. Yuan, Y., Gao, Y., Mao, L., Zhao, J., 2008. Optimisation of conditions for the
Dickinson, E., 1993. Towards more natural emulsifiers. Trends in Food Science & preparation of b-carotene nanoemulsions using response surface methodology.
Technology 4, 330–334. Food Chemistry 107, 1300–1306.