This article was downloaded by: [Nova Southeastern University]

On: 30 December 2014, At: 18:50

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Drying Technology: An International Journal

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/ldrt20

Optimization of Fish Oil Spray Drying Using a

Protein:Inulin System
a b b b
Diego Alvarenga Botrel , Soraia Vilela Borges , Regiane Victória De Barros Fernandes &
b
Eloá Lourenço Do Carmo
a
Institute of Agricultural Sciences , Federal University of Viçosa , Rio Paranaíba , Minas
Gerais , Brazil
b
Food Science Department , Federal University of Lavras , Lavras , Minas Gerais , Brazil
Published online: 24 Jan 2014.

Click for updates

To cite this article: Diego Alvarenga Botrel , Soraia Vilela Borges , Regiane Victória De Barros Fernandes & Eloá Lourenço Do
Carmo (2014) Optimization of Fish Oil Spray Drying Using a Protein:Inulin System, Drying Technology: An International Journal,
32:3, 279-290, DOI: 10.1080/07373937.2013.823621

To link to this article: http://dx.doi.org/10.1080/07373937.2013.823621

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Drying Technology, 32: 279–290, 2014
Copyright © 2014 Taylor & Francis Group, LLC
ISSN: 0737-3937 print/1532-2300 online
DOI: 10.1080/07373937.2013.823621
Optimization of Fish Oil Spray Drying Using a Protein:Inulin
System
Diego Alvarenga Botrel,1
2 Soraia Vilela Borges,2 Regiane Victória de Barros
Fernandes,2 and Eloá Lourenço do Carmo2
1
Federal University of Viçosa, Institute of Agricultural Sciences, Rio Paranaíba, Minas Gerais, Brazil
2
Federal University of Lavras, Food Science Department, Lavras, Minas Gerais, Brazil
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
The aim of this work was to investigate the use of an unconven-
tional polymeric material, inulin, in fish oil spray drying. The influ-
ence of inlet air temperature, oil load, and carrier substitution
(wheyproteinisolatebyinulin)onthephysicalpropertiesofparticles
containing fish oil was investigated. Increasing the oil load caused an
increase in the surface oil; however, the presence of inulin favorably
decreasedtheparticlesurfaceoilwhenahighoilladwasapplied.The
parameters hygroscopicity and wettability were also influenced by
theinulinfraction.Thehigherinletairtemperatureproducedlarger
particles with lower densities. The best operating conditions were
determined to be an air temperature of 185
C, a 40% inulin fraction,
and a 6% oil load. The results indicate that inulin is an alternative
carrier in the fish oil spray-drying process.
Keywords Encapsulation; Fructan; Omega-3; Response surface
methodology; Spray drying; Whey protein
INTRODUCTION
Microencapsulation has become an attractive technique
to convert liquid food components, such as oils and flavor-
ings, into a dry and free-flowing powder form, which is easy
to handle and incorporate into a dry food system.1
2 Fur-
thermore, microencapsulation is a highly appropriate process
for protecting heat-sensitive components, such as polyunsa-
turated oils, anthocyanins, and volatile flavorings.3−6
Spray drying is the most common technology for both
the food industry and the pharmaceutical industry as a
microencapsulation process.7 It is characterized by the
transformation of feed from a fluid state into a dried pro-
duct by spraying the feed into a hot drying medium.8 The
production of microencapsulated powders by spray drying
involves the formation of a stable emulsion in which the
wall material acts as a stabilizer for the core material and
provides the advantages of relative low temperatures and
very short particle residence times.9
10
Correspondence: Professor Diego Alvarenga Botrel, Federal
University of Viçosa, Institute of Agricultural Sciences, BR
354 – km 310, Rio Paranaíba, 38810-000, Minas Gerais, Brazil;
E-mail: diego.botrel@ufv.br
279
The quality of spray-dried products is dependent on the
operating parameters, such as the concentration of the wall
material, oil load, and process temperatures. In general, there
is an optimum core concentration that can be encapsulated
efficiently, which avoids the occurrence of surface oil and
ensures a high yield.10
11 The inlet air temperature (160–
220
C) leads to a rapid formation of the semi-permeable
membrane on the droplet surface, which contributes to a
higher retention. When the air inlet temperature is low, the
low evaporation rate causes the formation of microcapsules
with high-density membranes, high water content, and poor
fluidity as well as a tendency for agglomeration.12 However,
excessive temperatures could cause heat damage to the dry
product and surface imperfections.11
The nature and composition of the wall materials are also
important factors in the encapsulating process. Depending
on the core material and the characteristics desired in the
final product, wall materials can be selected from a wide var-
iety of natural and synthetic polymers. Because almost all
spray-drying processes in the food industry are conducted
from an aqueous feed formulation, the wall material must
be soluble in water at an acceptable level.7 In addition to
its high solubility, a wall material for microencapsulation
by spray drying should possess good emulsification, film-
forming and drying properties, and the wall-concentrated
solutions should have low viscosity.13
Carbohydrates, milk proteins, and new, emerging biopo-
lymers compose the three main classes of wall materials that
are generally available and suitable for spray-drying micro-
encapsulation.11 Carbohydrates lack active surface proper-
ties and must be chemically modified or used in conjunction
with emulsifying agents to encapsulate hydrophobic core
materials.10 Proteins, such as caseinates, gelatin, and whey
proteins, have functional properties as wall materials, which
include the ability to form film, interact with water, and sta-
bilize emulsion droplets.11
14 The incorporation of hydro-
lyzed carbohydrates into the wall system has been shown
to improve the drying properties of the wall matrix, prob-
ably by enhancing the formation of a dry crust around the
drying droplets and increasing the oxidative stability by
 280 BOTREL ET AL.
reducing oxygen permeability.15
16 Inulin, a mixture of
polysaccharides composed of fructose unit chains (linked
by -(2→1) D-fructosyl-fructose bonds) of various length
and terminated (generally) by a single glucose unit,17 is
classified as a prebiotic dietary fiber. In addition, inulin has
several described technological and functional proper-
ties,18 and may be an alternative encapsulating agent for
use in the food industry.
Microencapsulation of fish oil, which is a rich source
of long-chain polyunsaturated fatty acids such as omega-3,
can provide the convenience of a solid powder, with
reduced oxidation, and can be used in many different fin-
ished food products. Several carrier agents, such as modi-
fied starch,19 whey protein and dextrin,15 and modified
cellulose,20 have been used for the encapsulation of fish
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
oil. However, limited data on the use of inulin as a carrier
have been published.
The aim of this study is to examine the effect of inlet
air temperature, fish oil load, and the substitution of whey
protein isolate (WPI) by inulin, as the encapsulating agent,
on the physical properties and encapsulation efficiency of
spray-dried particles containing fish oil. Response surface
methodology was used to demonstrate the influence of each
variable.
MATERIALS AND METHODS
Materials
Fish oil (Sundown Naturals, Boca Raton, FL, USA)
was used as a core material and had a composition of
18% eicosapentaenoic acid (C20:5) and 12% docosahexae-
noic acid (C22:6) and no added antioxidants, according to
the supplier. Whey protein isolate (WPI) with a minimum
of 90% of protein (Alibra Ingredientes Ltda, Campinas, Bra-
zil) and inulin with a degree of polymerization higher than
10 (92% of inulin and 8% of glucose/fructose/sucrose)
(Orafti
GR, BENEO-Orafti, Tienen, Belgium) were used
as wall materials.
Experimental Design
Response surface methodology that applied a rotatable
central composite design was used to determine the influ-
ence of three independent variables that were considered
to affect the spray-drying process. The three independent
variables were inlet air temperature (
C) (X1), inulin frac-
tion (%) (X2), and oil load (%) (X3). The experimental
design consisted of 18 points, including four replications
at the central point (Table 1), and was conducted in a ran-
dom order. The response variables were moisture content,
solubility, hygroscopicity, wettability, bulk density, tapped
density, surface oil, and particle size. The regression models
were evaluated for each response variable.21 The non-
significant terms were eliminated, and the resulting
equations were tested for adequacy and fit by an analysis
of variance (ANOVA). The relationship between the
independent variables and the response variables was
calculated by the second-order polynomial (Eq. (1)):

k
k
Y = 0 + i=1 i Xi + i=1 ii Xi Xi

k−1
k
+ i=1 j=i+1 ij Xi Xj (1)
where Y is the predicted response, 0 is a constant, i is the
linear coefficient, ii is the squared coefficient, ij is the cross
product coefficient, and k is the number of factors.
The optimum conditions for the microencapsulation of
fish oil were determined by analyzing the results of variables
that were significantly affected by the experimental con-
ditions and applying Response Desirability Profiling using
Statistica software (ver. 8, Stat Soft. Inc., Tulsa, OK,
USA), according to methodology described by Derringer
and Suich.22
Preparation of Emulsions and Spray Drying
Wall material solutions were prepared by dissolving WPI
and inulin in distilled water for each formulation. The
solutions were prepared on the day before emulsification
and kept overnight at room temperature to ensure a full
saturation of the polymer molecules. Oil was progressively
added to the wall material solution while stirring at
3,500 rpm for 10 minutes using a rotor-stator blender
(IKA T18 Ultra-Turrax Basic, Wilmington, USA). The oil
load (%) in relation to emulsion varied according to the
experimental design (Table 1). The emulsion was used as
the feeding liquid in the spray-drying process. The percent-
age of solids (wall materials) used in the feed solution was
kept at 15% (w/w) in relation to emulsion for all treatments
and was within the wall material concentration recom-
mended for oil encapsulation (i.e., 10% to 30%).23 The
emulsions were dried using a spray drier (model MSD 1.0;
Labmaq do Brasil, Ribeirão Preto, Brazil) equipped with
a two-fluid nozzle atomizer. The inlet air temperature
varied according to experimental design (120
C to 190
C);
the rates of the feed flow and atomizing air were kept at
085 L · h−1 and 35 L · min−1 , respectively.
Characterization of the Particles
Moisture Content
The moisture content of the powders was determined
using the AOAC (2000) method.24 The powder weight loss
percentage (%) was determined after oven drying at 105
C
until a constant weight was obtained and the moisture
content (%) was calculated.
Surface Oil Content and Encapsulation Efficiency
The percentage of extractable oil (i.e., free oil or surface
oil) was determined using extraction with petroleum
 PROTEIN:INULIN SPRAY-DRYING SYSTEM 281

TABLE 1
Experimental design for the spray-drying tests using a rotatable central composite design
Coded variables Real variables
Inlet air Inulin Oil load (%) in
Trials X1 X2 X3 temperature (
C) fraction (%) relation to emulsion
1 −100 −100 −100 134 10 2.7
2 −100 −100 100 134 10 6.3
3 −100 100 −100 134 40 2.7
4 −100 100 100 134 40 6.3
5 100 −100 −100 176 10 2.7
6 100 −100 100 176 10 6.3
7 100 100 −100 176 40 2.7
8 100 100 100 176 40 6.3
9 −168 000 000 120 25 4.5
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014

10 168 000 000 190 25 4.5

11 000 −168 000 155 0 4.5
12 000 168 000 155 50 4.5
13 000 000 −168 155 25 1.5
14 000 000 168 155 25 7.5
15 000 000 000 155 25 4.5
16 000 000 000 155 25 4.5
17 000 000 000 155 25 4.5
18 000 000 000 155 25 4.5

ether.25
26 Spray-dried powder (2.0 g) was dispersed into
25 mL of solvent in a volumetric flask and shaken frequently
for 10 minutes without particle disruption. The dispersion
was then filtered through Whatman no. 1 filter paper. The
collected microparticles were rinsed with 15 mL of pet-
roleum ether three times and the filtrate solution containing
the extractable oil was transferred to a petri dish to evapor-
ate the organic solvent under room temperature. The per-
centage of surface oil of the particles was calculated based
on the ratio of the extracted oil (g) and the initial mass of
the particles (g). In this study, no total oil analysis was per-
formed, assuming that all of the initial oil was retained in
the powder because of being non-volatile and depositions
of the unbounded oil on the dryer wall and degradation
of oil during spray drying were ignored. Surface oil content
was selected criteria for encapsulation efficiency.25
27
28
Hygroscopicity, Wettability, and Solubility
The hygroscopicity was determined according to the
method proposed by Cai and Corke29 with some modifica-
tions. The powder samples of each treatment (approximately
1 g) were placed in a container with a saturated NaCl sol-
ution (75% RH) at 25
C. After two weeks, the samples
were weighed, and hygroscopicity was expressed as g of
adsorbed moisture per 100 g of dry solids (g · 100 g−1 ).
The wettability of the powders was determined using the
method of Fuchs et al.30 with some modifications. The
powder samples (0.1 g) were sprinkled over the surface of
100 mL of distilled water at 20
C without agitation. The
time it took until the last powder particles’ submerging
was recorded and used for a relative comparison of the
extent of wettability between the samples.
The solubility of the powders in cold water was evaluated
according to the method proposed by Eastman and Moore
(1984)31 with some modifications. Distilled water (20 mL)
was measured and transferred into a 50 mL beaker. A pow-
der sample (2.5 g) was weighed and added into the beaker
containing water and stirred at a low speed. After all the
sample was added, the stirring was switched to high speed
for two minutes. The material was then transferred to a
50 mL centrifuge tube and centrifuged at 760 × g for 15
minutes. A 5 mL aliquot of the supernatant was transferred
into a tared petri dish and dried in an oven at 110
C for 4 h.
The cold-water solubility (S) was calculated by Eq. (2):
grams of solids in supernatant × 4
S % = × 100% (2)
grams of sample
Bulk and Tapped Densities
The powders were gently loaded into a 100 mL tared grad-
uated cylinder until they reached the 100 mL mark; the grad-
uated cylinders were weighed. The volume, read directly from
the cylinder, was then used to calculate the bulk density (bulk )
according to the following relationship: mass/volume.32 For
 282 BOTREL ET AL.
tapped density (tapped ), approximately 5 g of powder was
freely poured into a 25 mL graduated glass cylinder, and the
samples were repeatedly tapped manually by lifting and
dropping the cylinder under its own weight at a vertical
distance of 10 cm until a negligible difference in volume
between succeeding measurements was observed. Given the
mass m and the apparent (tapped) volume V of the powder,
the powder tapped density was computed.8
Particle Size Distribution
The particle size distribution was measured using a
Mastersizer 2000 laser light diffraction instrument, model
Hydro 2000 MU (Malvern Instruments, Worcestershire,
UK). A small powder sample was suspended in isopropyl
alcohol under agitation, and the particle size distribution
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
was monitored during each measurement until successive
readings became constant. The volume-weighted mean diam-
eter (D[4,3]) was measured, and the particle size distribution
of the powder (span) was calculated as follows (Eq. (3)):
d 0
9 − d0
1
span = (3)
d0
5
where d(0,9), d(0,5), and d(0,1) are the equivalent
volume diameters at 90%, 50%, and 10% cumulative volume,
respectively.32
Particle Morphology
For the optimal conditions, particle morphology was
evaluated by scanning electron microscopy (SEM). The
powders were attached to a double-sided adhesive tape
mounted on SEM stubs with a diameter of 1 cm and a
height of 1 cm and then coated with gold in a vacuum and
examined with an MEV 1430 VP–LEO scanning electron
microscope (Electron Microscopy Ltd., Cambridge, UK).
The scanning electron microscope was operated at 20 kV with
a magnification of 900–1200×.
Moisture Sorption Isotherm
The sorption isotherms were determined by the gravi-
metric static method using a saturated saline solution at
25
C. Seven saturated saline solutions (NaCl, K2 CO3 ,
MgCl2 , LiCl, Mg(NO3 )2 , KCl, and NaNO3 ) were used with
water activity varying from 0.11 to 0.84. The data from the
moisture sorption isotherms were correlated with the water
activity using the following mathematical models: GAB,
Halsey, Henderson, Oswin, and Smith. The most suitable
model was chosen based on the high coefficient of determi-
nation (R2 ) and the low mean relative percentage deviation
modulus (E), defined in Eq. (4):
 
100
N mi − mpi 
E= (4)
N i=1 mi
where mi is the experimental value, mpi is the predicted
value, and N is the population of the experimental data.
RESULTS AND DISCUSSION
Response Surface Analysis
The results of microcapsules characterization are shown
in Table 2. The second-order regression coefficients for the
polynomial equation, values of the Fcal/Ftab ratios, and
determination coefficients (R2 ) were calculated (Table 3).
Non-significant (p > 010) terms were eliminated from the
equations. The models obtained for the variables’ tapped
density and particle size were non-predictive, presenting
R2 lower than 0.70. The other obtained models were
considered predictive, showing significant regression and
low residual values.
Microcapsules Characterization
All of the moisture content values encountered for the
studied treatments were below 2.0%, similar to those found
by Frascareli et al.33 (0.76%–3.23%) for the encapsulation
of coffee oil using gum Arabic and by Santana et al.34
(0.3%–1.9%) for spray drying of pequi pulp applying
temperatures ranging from 140
C to 200
C. The difference
between water content in the particles was greatly
influenced (p < 005) by the inlet air temperature. As the
temperature increased, there was an increase in the tem-
perature gradient between the air and the sprayed product,
resulting in a greater heat and mass transference that
increased the water evaporation rate. Similar results were
observed by Aghbashlo et al.35 in spray drying fish oil
using dairy proteins. However, the moisture content also
increased when a very high inlet temperature was applied.
This result was related to the increased rate of crust forma-
tion, which hinders the water diffusion from the sprayed
particles. An increased oil load led to a decrease in moisture
content, probably due the hydrophobicity of this compo-
nent. The inulin fraction also influenced this variable.
Low moisture content was observed with high WPI content
and an increase of inulin in the formulation. In the first case,
the protein ability to bind the water molecules makes it
difficult for the water to diffuse; in the second case, the
presence of inulin contributes to rapid crust formation
related to the shorter size of their chains, hindering the
water release. Goula and Adamapoulos36 observed
increased water content with an increased maltodextrin
concentration in the production of orange juice powder
and concluded that the presence of the longer maltodextrin
molecules hindered water diffusion. The water content was
not a critical parameter because all of the powders produced
had low moisture content and were within the range that is
considered safe for avoiding microorganism growth and the
development of product alteration related to the initial
water content.
 PROTEIN:INULIN SPRAY-DRYING SYSTEM 283

TABLE 2
Observed values for the studied response variables
Moisture Surface Hygroscopicity Wettability Solubility Bulk density Tapped density D[4.3]
Trials content (%) oil (%) (g · 100−1 ) (s) (%) (g · mL−1 ) (g · mL−1 ) ( m)
1 189 49 119 124 823 0208 0295 103
2 044 94 100 127 777 0213 0341 113
3 112 45 119 153 803 0232 0383 105
4 109 88 110 50 730 0221 0372 101
5 115 48 117 215 815 0170 0259 102
6 005 115 95 62 736 0193 0298 137
7 010 50 117 160 826 0178 0259 112
8 005 76 110 80 794 0224 0316 109
9 189 66 106 125 788 0201 0338 97
10 005 76 105 134 780 0196 0294 126
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014

11 153 84 116 160 783 0184 0295 136

12 065 68 108 58 782 0225 0378 98
13 049 24 111 253 842 0198 0317 98
14 015 139 91 40 719 0201 0292 156
15 030 64 103 185 791 0201 0321 96
16 035 71 109 175 783 0199 0322 106
17 030 75 103 185 777 0195 0322 110
18 035 72 101 155 782 0202 0324 111

The percentage of surface oil in the particles was nega-
tively (p < 005) influenced by inulin fraction and, to a
greater extent, positively influenced by the initial oil load
in the emulsion. In both cases, only the linear term had an
influence (Fig. 1). Therefore, higher values of oil load con-
tributed to an increase in the particles’ surface oil. At higher
oil loads, the wall material concentration may not be
sufficient to fully cover the emulsion oil droplets and thus
may reduce the encapsulation efficiency. Similar behavior
was observed by Minemoto et al.37 in the encapsulation
of linoleic acid and by Garcia et al.38 in the encapsulation
of basil essential oil, both using gum Arabic as the wall

TABLE 3
Coded second-order regression coefficients for the significant responses
Moisture Superficial Hygroscopicity Wettability Solubility Bulk density Tapped density D[4.3]
Coefficients content (%) oil (%) (g · 100−1 ) (s) (%) (g · mL−1 ) (g · mL−1 ) ( m)
0 0.31∗∗ 7.23∗∗ 10.49∗∗ 175.1∗∗ 78.51∗∗ 0.200∗∗ 0.318∗∗ 10.80∗∗
1 −046∗∗ ns ns ns ns −0009∗∗ −0024∗∗ 0.63∗∗
11 0.21∗∗ ns ns −171∗∗ ns ns ns ns
2 −019∗∗ −054∗∗ ns −188∗∗ ns 0.010∗∗ 0.020∗∗ −067∗∗
22 0.26∗∗ ns 0.37∗∗ −242∗∗ ns ns ns ns
3 −023∗∗ 2.75∗∗ −066∗∗ −506∗∗ −320∗∗ 0.005∗ ns 1.00∗∗
33 ns ns ns −110∗∗ ns ns ns 0.52∗
12 ns ns ns ns 1.68∗∗ ns ns ns
13 ns ns ns ns ns 0.009∗∗ ns ns
23 0.31∗∗ −054∗ 0.30∗ ns ns ns ns ns
Fcal/Ftab 8.01 13.72 4.45 3.96 23.97 3.35 3.61 1.74
R2 0.93 0.91 0.76 0.82 0.92 0.76 0.64 0.63
1-inlet air temperature; 2-inulin fraction; 3-oil load in relation to emulsion.
ns: not significant (p > 0.10).
∗∗
Significant at 5% probability.
∗
Significant at 10% probability.
 284
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
FIG. 1. Response surface for the surface oil variable at an inlet air
temperature of 155
C.
material. It was also verified that the interaction between
those two variables was significant (p < 010). When higher
oil loads were applied, the WPI substitution by inulin was
favorable for decreasing the surface oil. Inulin, a mixture
of short-chain molecules, may have hindered the escape of
oil from the particles, making them more compact. When
high concentrations of core material are used, a decrease
of the retention efficiency has been previously attributed
to the large amount of material near to the drying surface,
which creates a shorter diffusion path to the air interface
and particles.11
An increase in the oil load negatively influenced the
particle’s water adsorption characteristics; that is, when
higher oil load was applied, the powder’s hygroscopicity
was decreased (Fig. 2). This result is related to the presence
of larger surface oil quantity due to the higher oil load used
in the emulsion. In this case, the particle surface, in the
presence of non-encapsulated oil, presents a hydrophobic
characteristic, and thus the water adsorption is reduced.
This behavior was also observed in coffee oil encapsula-
tion.33 Moreover, WPI showed a positive influence on
the quadratic term. A minimum amount of water adsorp-
tion was estimated when intermediary values of inulin frac-
tion were used; however, there was a slight variation due to
the oil load. Water adsorption is an important factor during
powder storage because lipid oxidation and particle flow are
influenced. The values for hygroscopicity ranged from
9.1 g · 100 g−1 to 11.9 g · 100 g−1 . Spray-dried particles can
easily absorb water from the surrounding environment,
resulting in a sticky powder surface and caking unless
necessary precautions are taken.8
The powder’s solubility decreased when the initial oil
load was increased and was strongly influenced by the
emulsion oil load. This result is related to the particles’
hydrophobic characteristics, as powder dissolution is
BOTREL ET AL.
closely connected with the water affinity of its components.
In this case, a higher oil load and thus higher surface oil, as
already discussed, may decrease the solubility of the
powder. The encountered values ranged from 71.9% to
84.2%. An interaction between inulin fraction and inlet air
temperature was observed. Higher solubility values were
estimated when higher temperatures and the presence of
inulin were applied (Fig. 2). Inulin, as a component with
hydrophilic properties, contributes to an increase in the
powder solubility. This behavior is in accordance with the
results of Johnston and Duckworth39 and Favaro-
Trindade et al.,40 who reported that small, highly polar
molecules are readily soluble in water.
The ability of dry powders to mix with water is one of the
most important physical properties related to its reconstitu-
tion. The instantanization time, measured through the
powder wettability, relates to the capability of the powder
to dissolve in water; that is, the ability to rehydrate in
water.41 A longer instantanization time contributes to an
inefficient solubilization and a possible rejection by consu-
mers. Values varying from 40 s to 253 s were obtained for
the treatment at a temperature of 155
C, an inulin fraction
of 25%, and an oil load of 1.5% and 7.5%, respectively. In
studies on the encapsulation of vegetable oil42 and avocado
oil43 by spray drying, wettability times between 1140 s and
2220 s and between 457 s and 1083 s, respectively, were
observed. These observed times, which were longer than
those observed in the present study, may be attributed to
the higher amount of mass employed (1.0 g of particles
instead of 0.1 g per 100 mL of water) in the methodology,
which causes the time to be longer. The oil load most influ-
enced the variation of the wettability times. Specifically, a
higher oil load correlated with less time necessary for the
powders to become wet (Fig. 3). When hydrophilic powder
materials like whey protein and soluble carbohydrates are
added into the water, mainly when they did not pass
through an instantanization process, these materials tend
to wet too fast and form lumps that are wet on the outside
and dry on the inside, which hinder water spreading
through the particles that form these lumps. Thus, the pres-
ence of surface oil over the particles helped to reduce the
lumps’ formation by slowing down the wettability of the
powder just enough so that it disperses evenly into the sol-
ution. The wettability time presented a maximum estimated
value when an intermediary inlet air temperature (about
155
C) and inulin fraction (about 30%) were applied. The
use of high temperatures can contribute to a greater occur-
rence of lumps upon addition to water because of the
production of particles with a very dry surface.
Particles with bulk and tapped densities were strongly
influenced by the wall material composition; that is, the
WPI substitution by inulin. In addition, higher densities were
observed with a higher presence of inulin. Inulin molecules
are smaller than whey protein molecules, and because of this,
 PROTEIN:INULIN SPRAY-DRYING SYSTEM
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
FIG. 2. Surface responses for (a) hygroscopicity (%), inlet air temperature of 155
C; (b) solubility (%), oil load of 4.5%; (c) solubility (%), inulin fraction
of 25%.
there is a greater possibility of wall material accommodation
into the open spaces between molecules, which permits the
formation of more compact structures with higher densities.
Values between 0.170 g · mL−1 and 0.232 g · mL−1 for bulk
densities and between 0.259 g · mL−1 and 0.383 g · mL−1 for
tapped densities were observed. Densities varying from
0.52 g · mL−1 to 0.67 g · mL−1 were found for the encapsula-
tion of pigments.29 The higher values observed in the inves-
tigation of Cai and Corke,29 compared to the present work,
may be related to the application of only maltodextrin as a
wall material, which provides a reduced molecular structure
and produces more compact particles. A negative influence
for the linear term of inlet air temperature was verified. In
general, lower density values were obtained at higher tem-
peratures. High temperatures promote a rapid air expansion
into the particles and enable the crust to dry quickly, leading
to the formation of larger particles with larger voids in the
material’s interior.9 According to Reineccius,44 compared
to slower drying conditions, faster drying conditions
285
produce larger particles, create a rigid structure, and do
not allow particles to shrink during the drying process. A
more compacted bed (that is, higher density) results in a
lower probability of trapped air occurring in the structure
and consequently a lower probability of oxygen contact
with the oil.
The average particle size was influenced by the three
studied factors. Higher temperatures led to the production
of larger particles due to the greater expansion of the inter-
nal air in the particles. Some authors have suggested that
the large sizes produced from such drying conditions are
due to the rapid crust formation at the droplet surface,
which hinders the shrinkage of the material.35 The pres-
ence of inulin caused a reduction of the particle size, prob-
ably due to the smaller chain size of this component, which
allows a more compacted structure. The added oil, which
generally settles inside the carrier matrix, may allow greater
chain mobility and consequently facilitate particle expan-
sion. The magnitude of particle expansion, after the
 286 BOTREL ET AL.
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014

FIG. 3. Response surfaces for wettability (s): (a) oil load of 4.5%; (b) inlet air temperature of 155
C; (c) inulin fraction of 25%.

shrinkage that occurs at the beginning of the drying process,
is dependent on the drying rate and the viscoelastic proper-
ties of the material. Observed values for the particles’
volume mean diameter ranged from 9.6 m to 15.6 m.
Tonon et al.45 found particles’ mean diameter equal to
22.35 m in the encapsulation of flaxseed oil using whey
protein and 10% of oil concentration. The homogeneity of
the particle size distribution was measured through the
spread (span), where values below 3.5 were obtained, which
indicates high homogeneity.
Optimization Process
The variables used to determine the optimal process
conditions were those related to the powder reconstitution
(solubility, wettability, and hygroscopicity) and the percent-
age of surface oil in the particles. The moisture content,

FIG. 4. Contour curves for the optimal condition.

 PROTEIN:INULIN SPRAY-DRYING SYSTEM 287

FIG. 5. Scanning electron microscope visualization of the particles produced at the optimal conditions: inlet air temperature of 185
C, inulin fraction of
40%, and oil load of 6%.
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014

bulk and tapped densities, and particles size, despite being
significant and predictive, were not used in the optimization
process. Therefore, the optimal conditions were determined
based on a minimum of the surface oil and hygroscopicity
variables and a maximum of the solubility and wettability
for the particles.
The simultaneous experimental response evaluation
provided the contour curves for the values of the factors
that produced the optimal conditions (Fig. 4). Through
the evaluation of this function, the optimal measures
encountered for fish oil encapsulation in the WPI and inu-
lin matrix were an inlet air temperature of 185
C, an inulin
fraction of 40%, and a fish oil load of 6%. Applying these
optimal conditions, the values predicted for the response
variables are 78.2% of solubility, 45 s of wettability
time, 10.8% of hygroscopicity, and 8.6% of surface oil in
the particles.
Particle Characterization Under the Optimal Conditions
The particles produced under the optimal conditions were
characterized regarding the morphology and the moisture
adsorption, determined using scanning electron microscopic
visualization and moisture sorption isotherm, respectively.
The particles showed spherical forms and the occurrence
of small dents in some particles; however, most of the sur-
face was relatively smooth (Fig. 5). No evidence of cracks
or fissures was found. Sheu and Rosenberg16 also found
similar structures in the encapsulation of ethyl caprylate
using WPI and maltodextrin as wall materials and verified
that, with increasing proportions of WPI, the particle sur-
face becomes smoother and has no evidence of dents. The
occurrence of dents is probably due to shrinkage at the
early stages of the drying process, and the particles may
be smoother when favorably high drying rates and a wall
material with viscoelastic properties are used.16

FIG. 6. Moisture sorption isotherm (GAB model) for the powder produced under the optimal condition: inlet air temperature of 185
C, inulin fraction of
40%, and oil load of 6%.
 288 BOTREL ET AL.

Applying the optimal conditions, the moisture sorption
isotherm was also constructed. The GAB model was
chosen as the best model to describe the moisture sorption
isotherm behavior due the high determination coefficient
(R2 ) and the low mean relative deviation (%) (Table 4
and Fig. 6). The obtained isotherm is described as a sig-
moid type II.46 The same type of isotherm was obtained
by Zhou and Labuza47 and Foster et al.48 during their
studies of WPI powder.
The addition of 40% of inulin to the WPI formulation
did not cause changes in the isotherm behavior compared
to WPI alone, as discussed above. The monolayer moisture
content, which is related to the amount of water bound to
the active sites, was 0065 g · g−1 . The powder’s physical
structure was not dramatically altered as the moisture
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
temperature of the produced formulation was not exceeded,
even at the high relative moisture applied in this experiment
(aw = 0.84 at 25
C). These results reinforce the feasibility of
using WPI and inulin as carriers in the encapsulation
process through spray drying to maintain the product
integrity during storage.
CONCLUSION
The substitution of whey protein isolate by inulin (inulin
fraction) had a positive effect over several studied variables.
The presence of inulin favorably decreased the particle
surface oil when a high oil load was applied. The addition
of a low inulin concentration to the wall material formu-
lation reduced the water adsorption. Inulin also favored
particles with small sizes. Increasing the temperature

adsorption increased. In addition, a change in the powder

state was not observed, indicating that the glass transition contributed to the formation of larger particles and lower
densities. The optimal process conditions were determined
TABLE 4 to be an inlet air temperature of 185
C, an inulin fraction
Estimated values of coefficients and statistical parameters of 40%, and a fish oil load of 6%. Under these conditions,
used to evaluate the adequacy of the models for the powder the particles had a smooth surface without cracks.
produced under the optimal conditions: inlet air temperature The GAB model was used to explain the moisture sorption iso-
of 185
C, inulin fraction of 40%, and oil load of 6% therm behavior. The application of inulin as a wall material
for the fish oil encapsulation process using spray drying is a
WPI (60): feasible alternative, and the functional properties of this
Models Equation Inulin (40) fiber support its use in food products.
Xeq CKaw
GAB Xeq = 1−Kaw 1−Kaw +CKaw 
Xm 0.065
C 4.456 FUNDING
K 0.746 The authors thank FAPEMIG (Minas Gerais State
E (%) 14.794 Foundation for Research Development, Brazil) for the financial
R2 0.943 support.
 1/b
Halsey Xeq = a
ln aw
a −0.002
b 0.071 REFERENCES
E (%) 27.871 1. Fang, Z.; Bhandari, B. Encapsulation of polyphenols: A review.
Trends in Food Science & Technology 2010, 21(10), 510–523.
R2 0.619
 1/b 2. Gharsallaoui, A.; Roudaut, G.; Chambin, O.; Voilley, A.; Saurel, R.
Henderson Xeq = ln1−aw 
−a
a 2.502 Applications of spray-drying in microencapsulation of food
ingredients: An overview. Food Research International 2007, 40(9),
b 4.401 1107–1121.
E (%) 16.949 3. Chen, Q.; Zhong, F.; Wen, J.; McGillivray, D.; Quek, S.Y. Properties
R2 0,878 and stability of spray-dried and freeze-dried microcapsules
 b
co-encapsulated with fish oil, phytosterol esters, and limonene. Drying
Oswin Xeq = a aw
1−aw 
a 0.073 Technology 2013, 31(6), 707–716.
b 0.461 4. Syamaladevi, R.M.; Insan, S.K.; Dhawan, S.; Andrews, P.; Sablani,
E (%) 21.597 S.S. Physicochemical properties of encapsulated red raspberry (Rubus
R2 0.930 idaeus) powder: Influence of high-pressure homogenization. Drying
Technology 2012, 30(5), 484–493.
Smith Xeq = a + b log1 − aw  a 0.019
5. Bringas-Lantigua, M.; Expósito-Molina, I.; Reineccius, G.A.;
b −0.075 López-Hernández, O.; Pino, J.A. Influence of spray-dryer air tem-
E (%) 20.665 peratures on encapsulated mandarin oil. Drying Technology 2011,
R2 0.932 29(5), 520–526.
6. Fernandes, R.V.B.; Borges, S.V.; Botrel, D.A. Influence of spray dry-
Xeq : equilibrium moisture content (g · g−1 dry powder); ing operating conditions on microencapsulated rosemary essential oil
Xm : monolayer moisture content (g · g−1 dry powder); C, K: properties. Ciência e Tecnologia de Alimentos 2013, 33(1), 171–178.
model constants related to monolayer and monolayer properties; 7. Gouin, S. Microencapsulation: Industrial appraisal of existing
aw: water activity; a, b: model parameters; E: mean relative devi- technologies and trends. Trends in Food Science and Technology
ation modulus; R2 : coefficient of determination. 2004, 15(7–8), 330–347.
 PROTEIN:INULIN SPRAY-DRYING SYSTEM
8. Goula, A.M.; Adamopoulos, K.G. Effect of maltodextrin addition during
spray drying of tomato pulp in dehumidified air: II. Powder properties.
Drying Technology 2008, 26(6), 726–737.
9. León-Matínez, F.M.; Méndez-Lagunas, L.L.; Rodríguez-Ramírez, J.
Spray drying of nopal mucilage (Opuntia fícus-indica): Effects on
powder properties and characterization. Carbohydrate Polymers
2010, 81(4), 864–870.
10. Hogan, S.A.; McNamee, B.F.; O’Riordan, E.D.; O’Sullivan, M.
Microencapsulating properties of sodium caseinate. Journal of
Agricultural and Food Chemistry 2001, 49(4), 1934–1938.
11. Jafari, S.M.; Assadpoor, E.; He, Y.; Bhandari, B. Encapsulation
efficiency of food flavours and oils during spray drying. Drying
Technology 2008, 26(7), 816–835.
12. Mofidi, N.; Aghai-Moghadam, M.; Sarbolouki, M.N. Mass
preparation and characterization of alginate microspheres. Process
Biochemistry 2000, 35(9), 885–888.
13. Reineccius, G.A. Spray-drying of food flavors. In Flavor Encapsula-
tion; Reineccius, G.A.; Risch, S.J., Eds.; American Chemical Society:
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
Washington, DC, 1988; 55–66.
14. Bylaite, E.; Venskutons, P.R.; Mapbpieriene, R. Properties of
caraway (Curum carvi L.) essential oil encapsulated into milk
protein-based matrices. European Food Research and Technology
2001, 212(6), 661–670.
15. Kagami, Y.; Sugimura, S.; Fujishima, N.; Matsuda, K.; Kometani, T.;
Matsumura, Y. Oxidative stability, structure, and physical character-
istics of microcapsules formed by spray drying of fish oil with protein
and dextrin wall materials. Food Engineering and Physical Properties
2003, 68(7), 2248–2255.
16. Sheu, T-Y.; Rosenberg, M. Microestructure of microcapsules consist-
ing of whey proteins and carbohydrates. Journal of Food Science 1998,
63(3), 491–494.
17. Ronkart, S.; Blecker, C.; Fougnies, C.; Van Herck, J.C.; Wouters, J.;
Paquot, M. Determination of physical changes of inulin related to
sorption isotherms: An X-ray diffraction, modulated differential scan-
ning calorimetry and environmental scanning electron microscopy
study. Carbohydrate Polymers 2006, 63(2), 210–217.
18. Franck, A.; Bosscher, D. Inulin. In Fiber Ingredients: Food Applica-
tions and Health Benefits; Cho, S.S.; Samuel, P., Eds.; CRC Press:
Boca Raton, FL, 2009; 41–60.
19. Serfert, Y.; Drusch, S.; Schmidt-Hansberg, B.; Kind, M.; Schawarz,
K. Process engineering parameters and type of n-octenylsuccinate-
derivatised starch affect stability of microencapsulated long chain
polyunsaturated fatty acids. Journal of Food Engineering 2009,
95(3), 386–392.
20. Kolanowski, W.; Laufenberg, G.; Kunz, B. Fish oil stabilization by
microencapsulation with modified cellulose.International Journal of
Food Science and Nutrition 2004, 55(4), 333–343.
21. Rodrigues, M.I.; Iemma, A.F. Experimental Design and Process
Optimization; Ed. Casa do Espirito Amigo Fraternidade Fe e Amor:
Campinas, SP, Brazil. Translation: Evan Michael Visser. Revision:
Hillary Castle de Menezes, 2012.
22. Derringer, G.; Suich, R. Simultaneous optimization of several
response variables. Journal of Quality Technology 1980, 12(4),
214–219.
23. Rosenberg, M.; Young, S.L. Whey protein as microencapsulating
agents, microencapsulation of anhydrous milkfat: Structure evalu-
ation. Food Structure 1993, 12(1), 31–41.
24. AOAC. Official Methods of Analysis, 17th ed.; Association of Official
Analytical Chemists: Gaithersburg, MD, 2007.
25. Jafari, S.M.; Assadpoor, E.; Bhandari, B.; He, Y. Nano-particle
encapsulation encapsulation of fish oil by spray drying. Food Research
International 2008, 41(2), 172–183.
26. Garcia, E.; Gutiérrez, S.; Nolasco, H.; Carreón, L.; Arjona, O.
Lipid composition of shark liver oil: Effects of emulsifying and
289
microencapsulation process. European Food Research and Technology
2006, 222(5–6), 697–701.
27. Jafari, S.M.; He, Y.; Bhandari, B. Role of powder particle size on
the microencapsulation efficiency of oil during spray drying. Drying
Technology 2007, 25(6), 1091–1099.
28. Augustin, M.A.; Sanguansri, L.; Bode, O. Maillard reaction products
as encapsulants for fish oil powders. Food Engineering and Physical
Properties 2006, 71(2), 25–32.
29. Cai, Y.Z.; Corke, H. Production and properties of spray-dried Amar-
anthus betacyanin pigments. Sensory and Nutritive Quality of Foods
2000, 65(6), 1248–1252.
30. Fuchs, M.; Turchiuli, C.; Bohin, M.; Cuvelier, M.E.; Ordonnaud, C.;
Peyrat-Maillard, M.N.; Dumoulin, E. Encapsulation of oil in powder
using spray drying and fluidised bed agglomeration. Journal of Food
Engineering 2006, 75(1), 27–35.
31. Eastman, J.E.; Moore, C.O. Cold Water Soluble Granular Starch for
Gelled Food Composition, U.S. Patent 4465702, 1984.
32. Jinapong, N.; Suphantharika, M.; Jamnong, P. Production of instant
soymilk powders by ultrafiltration, spray drying and fluidized bed
agglomeration. Journal of Food Engineering 2008, 84(2), 194–205.
33. Frascareli, E.C.; Silva, V.M.; Tonon, R.V.; Hubinger, M.D. Effect of
process conditions on the microencapsulation of coffee oil by spray
drying. Food and Bioproducts Processing 2012, 90(3), 413–424.
34. Santana, A.A.; Kurozawa, L.E.; Oliveira, R.A.; Park, K.J. Influence of
process conditions on the physicochemical properties of pequi powder
produced by spray drying. Drying Technology 2013, 31(7), 825–836.
35. Aghbashlo, M.; Mobli, H.; Madadlou, A.; Rafiee, S. Influence of wall
material and inlet drying air temperature on the microencapsulating of
fish oil by spray drying. Food and Bioprocess Technology 2013, 6(6),
1561–1569.
36. Goula, A.M.; Adamopoulos, K.G. A new technique for spray drying
orange juice concentrate. Innovative Food Science & Emerging Tech-
nologies 2010, 11(2), 342–351.
37. Minemoto, Y.; Hakamata, K.; Adachi, S.; Matsuno, R. Oxidation of
linoleic acid encapsulated with gum arabic or maltodextrin by spray-
drying. Journal of Microencapsulation 2002, 19(2), 181–189.
38. Garcia, L.C.; Tonon, R.V.; Hubinger, M.D. Effect of homogenization
pressure and oil load on the emulsion properties and the oil retention
of microencapsulated basil essential oil (Ocimum basilicum L.). Drying
Technology 2012, 30(13), 1413–1421.
39. Johnston, K.A.; Duckworth, R.B. The influence of soluble compo-
nents on water sorption hysteresis. In Properties of Water in Foods
in Relation to Quality and Stability; Simatos, D.; Multon, J.L., Eds.;
Martinus-Nijhoff Publishers: Dordrecht, The Netherlands, 1985;
65–82.
40. Favaro-Trindade, C.S.; Santana, A.S.; Monterey-Quintero, E.S.;
Trindade, M.A.; Netto, F.M. The use of spray drying technology to
reduce bitter taste of casein hydrolysate. Food Hydrocolloids 2010,
24(4), 336–340.
41. Gaiani, C.; Scher, J.; Ehrhardt, J.J.; Linder, M..; Schuck, P.; Desobry,
S.; Banon, S. Relationships between dairy powder surface compo-
sition and wetting properties during storage: Importance of residual
lipids. Journal of Agricultural and Food Chemistry 2007, 55(16),
6561–6567.
42. Turchiuli, C.; Fuchs, M.; Bohin, M.; Cuvelier, M.E.; Ordonnaud, C.;
Peyrat-Maillard, M.N.; Dumoulin, E. Oil encapsulation by spray
drying and fluidised bed agglomeration. Innovative Food Science &
Emerging Technologies 2005, 6(1), 29–35.
43. Bae, E.K.; Lee, S.J. Microencapsulation of avocado oil by spray dry-
ing using whey protein and maltodextrin. Journal of Microencapsula-
tion 2008, 25(8), 549–560.
44. Reiniccius, G.A. Multiple-core encapsulation: The spray drying of
food ingredients. In Microencapsulation of Food Ingredients; Vilstrup,
P., Ed.; Leatherhead Publishing: Surrey, 2001; 151–185.
 290 BOTREL ET AL.

45. Tonon, R.V.; Pedro, R.B.; Grosso, C.R.F.; Hubinger, M.D.
Microencapsulation of flaxseed oil by spray drying: Effect of oil
load and type of wall material. Drying Technology 2012, 30(13),
1491–1501.
46. Roos, Y. H. Phase Transition in Foods; Academic Press, Inc.: San
Diego, CA, 1995; 49–107.
Downloaded by [Nova Southeastern University] at 18:50 30 December 2014
47. Zhou, P.; Labuza, T.P. Effect of water content on glass transition and
protein aggregation of whey protein powders during short-term
storage. Food Biophysics 2007, 2(2–3), 108–116.
48. Foster, K.D.; Bronlund, J.E.; Paterson, A.H.J. The prediction of
moisture sorption isotherms for dairy powders. International Dairy
Journal 2005, 15(4), 411–418.