7/25/2019 Aguilar 2015

1/8

Protein adsorption onto alginate-pectin microparticles and filmsproduced by ionic gelation

Karla Crdova Aguilar a, Fernando Tello b, AndreaC.K. Bierhalz c, Ma.G. Garnica Romo d,Hctor E. Martnez Flores e, Carlos R.F. Grosso b,

a Programa Institucional de Maestra en Ciencias Biolgicas de la Universidad Michoacana de San Nicols de Hidalgo, Morelia, Michoacn, Mexicob Department of Foods and Nutrition, Faculty of Food Engineering, University of Campinas, Campinas, So Paulo 13083-862, Brazilc School of Chemical Engineering, University of Campinas, Campinas, So Paulo 13083-852, Brazild Facultad de Ingeniera Civil, Universidad Michoacana de San Nicols de Hidalgo, Morelia, Michoacn, Mexicoe Facultad de Qumico Farmacobiologa, Universidad Michoacana de San Nicols de Hidalgo, Morelia, Michoacn, Mexico

a r t i c l e i n f o

Article history:

Received 6 May 2014

Received in revised form 2 December 2014

Accepted 26 December 2014

Available online 2 January 2015

Keywords:

Ionic gelation

Electrostatic adsorption

Microparticles

Edible films

Pectin

Alginate

a b s t r a c t

Microparticles and films containing sunflower oil were produced by ionic gelation using a 1:1 algi-

nate:pectin mixture and were electrostatically coated with whey and egg white proteins. Emulsions of

the polysaccharide mixture and the protein solutions were evaluated in terms of their zeta potentials.

The microparticles were characterized based on their mean size, size distribution, moisture content, cal-

cium content, adsorbed protein content, encapsulation efficiency and morphology. The films were char-

acterized with respect to their thickness, moisture content, calcium content, adsorbed protein content,

mechanical properties, water vapor permeability and morphology. High encapsulation efficiency

(87.6% at pH 3.5 and 90.8% at pH 3.75) was obtained for the microparticles produced by ionic gelation.

The calcium content after ionic gelation was significantly higher in the films (on average, 3.0lmol/mg db) than in the microparticles (on average, 1.62 lmol/mg db). For the microparticles, an increase inthe protein content in solution yielded an increase in the protein content adsorbed, independent of the

type of protein used. When 4% protein in solution was used, protein adsorption onto the microparticles(59.2% for whey protein and 45.5% for egg white protein) was significantly higher than that onto thefilms

(25.3% for whey protein and 24.1% for egg white protein) likely due to the smaller amount of calcium

present on the microparticles and the larger surface area of the particles relative to that of the films.

Although the process of producing films by ionic gelation and later coating them with proteins was

straightforward, homogeneous drying of the films was difficult.

2015 Elsevier Ltd. All rights reserved.

1. Introduction

External ionic gelation is a simple process used to obtain gels in

which, an anionic polysaccharide solution is dripped over an ionic

solution at appropriate concentrations, making it possible to obtain

gels of different shapes and sizes that form three-dimensional

structures with a high water content (Gombotz and Wee, 1998;

Smrdel et al., 2008).

Among polysaccharides, alginate, pectin or a mixture of the two

may be used to produce matrices in the form of microparticles or

films. The cation concentration, the ionic strength and the pH

determine the kinetics of gel formation as well the volume and sta-

bility of gel beads (Mestdagh and Axelos, 1998).

Alginate has a linear structure and a high molar mass and com-

prises two types of uronic acids, b-D mannuronic acid (M) and a-Lguluronic acid (G), with either homopolymeric or heteropolymeric

blocks in which the G units form crosslinks with divalent ions, to

produce egg-box model gels (Grant et al., 1973). Alginate has

pKa values between 3.20 and 3.38 (Martinsen et al., 1992).

The characteristic structure of pectin is a linear chain composed

ofa (1? 4)-D-galacturonic acids that are partially esterified withmethoxyl groupsand neutral sugars, such as galactose, glucose,

rhamnose, arabinose and xylose. Pectin has a pKa value of 2.9

(Ralet et al., 2001).

Although ionic gelation is a simple and mild technique, the gels

produced are porous, which can accelerate oxygen permeation

through the matrix or allow for the release of active compounds

with low molar mass that are inserted into the gels (Sezer and

Akbuga, 1999). The size of the pores varied greatly, ranging from

http://dx.doi.org/10.1016/j.jfoodeng.2014.12.020

0260-8774/2015 Elsevier Ltd. All rights reserved.

Corresponding author at: Depan, Fea, Unicamp. Tel.: +55 19 3521 4079.

E-mail address: grosso@fea.unicamp.br(C.R.F. Grosso).

Journal of Food Engineering 154 (2015) 1724

Contents lists available at ScienceDirect

Journal of Food Engineering

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

http://dx.doi.org/10.1016/j.jfoodeng.2014.12.020mailto:grosso@fea.unicamp.brhttp://dx.doi.org/10.1016/j.jfoodeng.2014.12.020http://www.sciencedirect.com/science/journal/02608774http://www.elsevier.com/locate/jfoodenghttp://www.elsevier.com/locate/jfoodenghttp://www.sciencedirect.com/science/journal/02608774http://dx.doi.org/10.1016/j.jfoodeng.2014.12.020mailto:grosso@fea.unicamp.brhttp://dx.doi.org/10.1016/j.jfoodeng.2014.12.020http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://crossmark.crossref.org/dialog/?doi=10.1016/j.jfoodeng.2014.12.020&domain=pdfhttp://-/?-

7/25/2019 Aguilar 2015

2/8

5 to 200 nm (Mestdagh and Axelos, 1998; Smidsrod, 1974). To cir-

cumvent this limitation and improve the functionality of the gels in

film or microparticle form, the literature proposes a mixing with

other biopolymers (Devi and Kakati, 2013) or coating with a layer

of polyelectrolytes with the opposite charge by electrostatic com-

plexation (Gbassi et al., 2011; Hbrard et al., 2010).

Complexes between proteins and polysaccharides can form

spontaneously in an aqueous solution due to the electrostatic

interactions between groups with opposite charges (Tolstoguzov,

2003). The formation and stability of the complexes are affected

by the pH, ionic strength, protein: polysaccharide ratio, and bio-

polymer charge density as well as production conditions such as

temperature, shearing and agitation time (Ye, 2008). The formation

of protein and polysaccharide complexes generally occurs between

the pKavalues of the anionic groups of the polysaccharide and the

isoelectric point (pI) of the protein (De Kruif et al., 2004).

Coating pectin and alginate microparticles with polycations

(such as chitosan and whey proteins) has been observed to result

in the formation of a membrane on the microparticlessurface

which retard the release of the active encapsulated compounds

(Humblet-Hua et al., 2011). The electrostatic interactions in the

microparticles can be controlled by varying the charge of the bio-

polymer surface. Recently, the adsorption of whey proteins onto

alginate or pectin microparticles was investigated (Gbassi et al.,

2011).

Egg white and whey proteins are two sources of globular pro-

teins used in food products. The major egg white protein is ovalbu-

min (54%, 44.5 kDa) with a reported isoelectric point of 4.8

(Oakenfull et al., 1997). Whey proteins contain a mixture ofb-lac-

toglobulin (82%, 18.5 KDa) anda-lactalbumin (15%, 14.5 KDa) withisoelectric points ranging from 4.4 to 5.2 (Damodaran, 2008).

The preparation of microparticles via external ionic gelation has

been widely studied and demonstrated to involve a high ease of

production (De Vos et al., 2010; Sriamornsak and Kennedy,

2011). However, considering that the external ionic gelation pro-

cess is very rapid and non-homogeneous, optimization of the film

production process remains a challenge.The objectives of this study were to produce microparticles con-

taining sunflower oil by ionic gelation using a mixture of alginate

and pectin (1:1, w/w). The microparticles were covered with egg

white and whey proteins in solution (1.7%, 3.0% and 4.0% protein

in solution). The zeta potentials of the proteins and polysaccharide

mixture solutions/emulsions were then evaluated at various pH

values. The microparticles were characterized in terms of their

morphology, size, size distribution and quantity of adsorbed

protein. Subsequently, films (FP) were produced from the algi-

nate-pectin mixture by ionic gelation. Then, the moist films were

covered with proteins at the highest levels of protein adsorptions

obtained with the microparticles. The amounts of calcium bound

to the microparticles and to the films produced by ionic gelation

were determined. The mechanical properties (tensile strengthand elongation at break), morphology, permeability to water vapor

and quantity of adsorbed protein of the films were evaluated.

2. Materials and methods

2.1. Materials

High molecular weight sodium alginate (ALG) with a high

guluronic acid content (lot G3512301 MANUGEL DMB; FMC

Biopolymer, Campinas, So Paulo, Brazil, 0.5 0.0% moisture con-

tent, AOAC, 2006), citrus pectin (PEC) GENU with a low amidation

and low methoxyl content [CP Kelco, Limeira, So Paulo, Brazil;

81.3 1.2% galacturonic acid content, 30.4 1.6% degree of esteri-fication and 10.1 1.0% degree of amidation, determined according

toFAO (2009)and 0.4 0.0% moisture content,AOAC, 2006], whey

protein concentrate (WPC) [Lacprodan lot Lac804U17601, 76,

Arla Foods Ingredients, Portea, Province of Crdoba, Argentina;

7.2 0.3% moisture content, 80.5 0.3% protein content and

3.9 0.0% ash content, determined according to AOAC (2006),

and 7.6 0.3% lipid content, determined according to Bligh and

Dyer (1959)], egg white proteins (OVA) (Saltos Alimentos LTDA,

Distrito Industrial, Parque do Lago Salto, So Paulo, Brazil;

6.7 0.5% moisture content, 90.5 0.8% protein content,

0.4 0.1% lipid content and 5.0 0.1% ash content), anhydrous cal-

cium chloride (Dinmica, Diadema SP, Brazil), sodium hydroxide

(Nuclear, Diadema SP, Brazil), hydrochloric acid and glycerol

(Merck, So Paulo SP, Brazil) and sulfuric acid (Synth, Diadema

SP, Brazil), commercial sunflower oil, and all other reagents used

were of analytical grade. All aqueous solution/emulsions were pre-

pared with destilled and deionized water.

2.2. Zeta potential of the biopolymers

The zeta potential was determined for protein solutions (OVA,

OVA + Glycerol, WPC and WPC + Glycerol) and for alginate-pectin

mixture solution and emulsion, at concentrations of 0.2% (w/w).

To prepare the polysaccharide mixture solution, pectin and algi-

nate were weighed as dry powders and mixed, deionized water

added to adjust the volume. All solutions were stirred overnight

at room temperature before measurements. The alginate-pectin

mixture solutions (2% w/w, moist weight basis) were emulsified

with 1.0% (w/w, moist weight basis) sunflower oil, at room temper-

ature, in a Turrax agitator (IKA, Works do Brasil, Rio de Janeiro RJ,

Brazil) at 14,000 rpm for 3 min. The resulting emulsion was diluted

to a concentration of 0.2% (v/v), and their zeta potentials were

measured, at varyingthe pH values ranging from 3.0to 7.0, at room

temperature. Before the readings, the pH of the solutions/emul-

sions was adjusted manually with HCl (0.1 N) or NaOH (0.1 N).

The measurements were performed using a Zetasizer Nano-Z (Mal-

vern, Worcestershire, U.K.). At least five measurements were

recorded at each pH.

2.3. Production of microparticles

The emulsion (alginate-pectin-sunflower oil) was sprayed over

a 2% (w/w) calcium chloride solution (pH adjusted to 3.5 or 3.75)

at room temperature using a double fluid atomizer (/, 1 mm) with

a distance of 12 cm between the tip of the atomizer and the surface

of the calcium chloride solution, an air pressure of 0.125 kgf/cm2

and a spraying rate of 555 mL/h. During spraying, the emulsion

was constantly agitated at room temperature. After spraying, the

microparticles were kept in calcium chloride solution for 30 min.

Next, the microparticles were washed three times with deionized

water (pH 3.5 or 3.75) and separated by sieves (/ 125lm). The

microparticles obtained by ionic gelation were transferred to eggwhite (pH 3.5) and whey (pH 3.75) protein solutions at concentra-

tions of 1.7%, 3% and 4% (w/w), using 50 g of moist particles and

200 mL of protein solution. The microparticles were kept in the

protein solutions for 30 min at room temperature under agitation

(500 rpm). Then, the microparticles were washed three times with

deionized water (pH 3.5 or 3.75) and separated by sieves (/

125lm) to remove any protein that was not adsorbed onto themicroparticles. Three sets of microparticles were produced inde-

pendently for each evaluated concentration. A fraction of the moist

microparticles was frozen and lyophilized (Mod. 501, Edwards

Pirani, Crawley, West Sussex, UK) at 40 C and 0.1 mmHg pres-

sure for a total cycle time of 48 h. The dry microparticles were

packed into flasks with lids and kept refrigerated. The protein

and dry material contents of the microparticles were determinedaccording to the AOAC (2006) guidelines using a conversion factors

18 K.C. Aguilar et al./ Journal of Food Engineering 154 (2015) 1724

7/25/2019 Aguilar 2015

3/8

of 6.68 and 6.38 for the amounts of nitrogen in OVA and WPC,

respectively. The nitrogen content of the alginate-pectin micropar-

ticles was determined and used to correct the total nitrogen con-

tent of the microparticles containing adsorbed protein. The

protein content was expressed as the weight percentage (%) of pro-

tein/weight of total solids on a dry weight basis.

2.4. Production of the alginate-pectin films

The emulsion (pectin-alginate sunflower oil) was prepared as

described above. The films were fabricated in two steps; cross-

linking and hardening, as in the production of microparticles. In

the first step, the film-forming emulsion (50 mL) was transferred

to a square acrylic plate (225 cm2) and sprayed for 1 h with a 2%

(w/w) calcium chloride solution with the pH adjusted (3.5 or

3.75) using a sprayer, with an air pressure of 1.0 kgf/cm2. During

spraying, the plate was kept inside a closed plastic chamber. After

spraying, the plate was removed from the chamber and kept at

room temperature for 15 min to stabilize the film. In the second

step (hardening time), the film was manually inverted and

100 mL of 2% (w/w) calcium chloride solution was added for 1 h

to complete the cross-linking and produce polysaccharide films

(FP3.50 and FP3.75). The moist films were cut into small discs (

29 mm, 0.5 g), and one disc was transferred to egg white (pH

3.5) and to whey (pH 3.75) protein solutions (10 mL) containing

glycerol (2:1, protein:glycerol, w/w) at a 4% protein concentration

and kept for 30 min under agitation at room temperature to pro-

duce protein-coated polysaccharide films (FP + OVA3.50 and FP +

WPC3.75). The films were then washed three times with deionized

water (pH 3.5 or 3.75) to remove any protein that was not

adsorbed onto the films. Three sets of films were independently

produced. The moist washed films were weighed to determine

the total solids and adsorbed protein following the same procedure

described previously for the microparticles. The films were dried

for 36 h at room temperature (25 3 C) and conditioned into des-

iccators (5054% RH) for three days before physical and mechani-

cal characterization. The preparation of the films is shown in theSupporting Materials (video data).

2.5. Characterization of the microparticles and films

2.5.1. Morphology of the microparticles and films

The morphologies of the moist microparticles with and without

protein coatings were examined using an optical microscope

(JENAVAL, Carl Zeiss, Toronto, Canada) with 12.5 and 25 objec-

tives and 1 and 1.25 optovar. Image capture was performed

using the EDN-2 Microscopy Image Processing System software

program. The lyophilized microparticles and dry films with and

without protein coatings were examined with a scanning electron

microscope (model JMS T300 Jeol, Tokyo, Japan). The samples

were fixed onto stubs with two-sided copper metallic tape andcovered with a fine layer of gold (180 s and 40 mA current) using

a Baltzer evaporator (Baltec SCD50, Liechtenstein). The analysis

conditions included voltage accelerations of 15 and 20 kV.

2.5.2. Determination of calcium present in the microparticles and films

after ionic gelation

The calcium content in the alginate-pectin microparticles and

films after ionic gelation was determined using an atomic absorp-

tion spectrophotometer (Analytik Jena AG-NOVAA300, Jena,

Germany) and a standard calcium solution (1000lg/mL, SCPScience, lot S120221015, Quebec, Canada) in absorption mode with

an air-acetylene flame detector. Samples of moist microparticles

and films (400 mg) were dissolved in 25 mL of 3% (w/w) sodium

citrate solution. Preliminary tests revealed that the sodium citratesolution did not affect the determination of the calcium content. To

avoid interference, microparticles and films were produced with-

out the addition of sunflower oil to evaluate the calcium content.

2.5.3. Average size, size distribution, moisture content of

microparticles with and without protein and efficiency of

encapsulation

The average size (d 0.5, lm) and the size distribution (d 0.1; d

0.9 and polydispersity index) of the moist microparticles weredetermined using a Mastersizer 2000 apparatus (Malvern, Worces-

tershire, WR, UK) with a Hydro 2000S sampling unit (Malvern,

Worcestershire, WR, UK). Deionized water (at pH 3.5 and 3.75)

was used as the dispersant material to read the samples. The mois-

ture content was gravimetrically determined by drying the sam-

ples at 105 C in a vacuum oven (Lab-Line, Squaroid, USA) for

24 h. After this period, the samples were weighed at 1-h intervals

using an analytical balance (Analytical Plus, Ohaus) until a con-

stant weight was reached. The encapsulation efficiency was only

determined after ionic gelation. Sodium citrate was added to 5 g

of moist microparticles at a concentration of 3% (w/w) to release

the oil, which was then quantified (Bligh and Dyer, 1959). The

encapsulation efficiency was determined on dry basis using the fol-

lowing equation:

EE% Total oil of the particlesg=total solidsg

Initial oilg=total solidsg 100

2.6. Functional characterization of the alginate-pectin films

The thickness of the films was measured with a digital microm-

eter (0.001 mm, model MDS-25S, Mitutoyo, Japan) at ten random

positions. The moisture content of the film was gravimetrically

determined as previously described. The Tensile strength (TS)

and elongation at break (E) of the preconditioned films were deter-

mined at room temperature using a TA.XT2 (Stable Microsystems

SMD, Surrey, UK) according to the ASTM standard method D882

(1995a)and the measurements were repeated at least five times.The water vapor permeability coefficient (WVP) through the films

was determined gravimetrically at 25 C (1 C) according to ASTM

method E96-95 (1995b).

2.7. Statistical analysis

Significant differences between the means obtained were eval-

uated using ANOVA and Tukeys test with the aid of the SAS pro-

gram at a 5% significance level. Each type of microparticle or film

was produced in three independent lots, and the determinations

were performed in triplicate for each lot.

3. Results and discussion

3.1. Zeta potential of the biopolymers

The 1:1 alginate-pectin mixture solution and emulsion had neg-

ative zeta potential values within the pH range tested (Fig. 1). The

electronegativity of the solution and emulsion increased with

increasing pH, from 31.9 4.3 mV and 30.7 4.5 mV (pH 3.5)

to 63.2 4.7 mV and 60.5 4.8 mV (pH 7), respectively. The

solution and emulsion showed similar values that were not signif-

icantly different, indicating that the addition of the oil did not alter

the net charge of the alginate-pectin mixture. Similar behavior was

previously reported for an alginate solution whose the zeta poten-

tial decreased from 8.7 to 68.4 mV when the pH was increased

from 2 to 8 (Harnsilawat et al., 2006). The zeta potential of the

WPC solution varied from 22.5 3.7 mV (pH 3) to 27.2 4.2 mV(pH 7), whereas the zeta potential of the OVA solution varied from

K.C. Aguilar et al. / Journal of Food Engineering 154 (2015) 1724 19

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-

7/25/2019 Aguilar 2015

4/8

28.5 3.7 mV (pH 3) to 27.9 5.3 mV (pH 7). The addition of

glycerol did not significantly alter the zeta potential of the pro-

teins. When WPC and WPC with glycerol solutions were evaluated,

the isoelectric point observed for both was 4.3. The egg white pro-

teins solutions with and without glycerol both had an isoelectric

point of 4.5, with no significant difference between the two.After ionic interaction between carboxyl groups from the

polysaccharide mixture and calcium ions, the zeta potential of

the particles or films is expected to be less electronegative than

the zeta potential of the polysaccharide mixture solution before

the addition of calcium ions. Opanasopit et al. (2008) obtained a

zeta potential of30 mV for a 0.1% pectin solution (low-ester ami-

dated pectin) at pH 4.0 and a zeta potential of10.4 mV for pectin

microparticles (particle size

7/25/2019 Aguilar 2015

5/8

egg white proteins (Fig. 3E) shows a surface similar to that of the

uncoated film but with lipid drops that are more spread out. In

general, the inside of the films with the protein coating ( Fig. 3D

and F) showed a more dense and compact structure relative to that

of the films without the protein coating (Fig. 3B).

3.5. Adsorption of proteins onto alginate-pectin microparticles and

films

The positively charged groups in the proteins can electrostati-

cally interact with negatively charged groups present in the poly-

saccharides and the amount of protein adsorbed onto the

microparticles or films is maximized when the charges of the

adsorbed protein are neutralized. Aside from electrostatic interac-

tions other interactions can occur, including Wan der Waals

forces, hydrogen bonds, hydrophobic effects and steric hindrance

and all can also be responsible for associations between protein

protein and protein surface interactions (Haynes and Norde,

1994).

The amount of protein adsorbed onto the coated microparticles

increased as the protein concentration in solution increased, inde-

pendent of the type of protein. The amount adsorbed in the case of

whey proteins (pH 3.75) were 38.8 4.7%, 46.1 2.1% and59.2 1.9%, whereas when the egg white proteins (pH 3.50) were

used, the amounts adsorbed were 25.7 1.4%, 33.1 2.2% and

45.5 2.8% at concentrations of 1.7%, 3% and 4% protein in solution,

respectively. Recently,Souza et al. (2012) studied whey proteins

adsorption onto pectin-based microcapsules produced by ionic

gelation and obtained 50% protein adsorption (w/w, db) when

working with a high protein concentration in solution (12%). The

adsorption of whey proteins and egg white proteins onto alginate

and pectin-based microparticles was previously evaluated (Tello

et al., 2015). The authors found that more protein was adsorbed

onto pectin microparticles (62% of protein adsorption egg white

proteins and 60% of protein adsorption whey proteins) than

onto alginate microparticles when 8% protein in solution was used.

Because the microparticles or films produced by ionic gelationare porous, it is possible that in addition to electrostatic interac-

tions, hydrogen bonds and hydrophobic protein-polysaccharide

interactions, protein molecules may have also penetrated the pores

of the microparticles or films. In the case of microparticles, more

whey protein than egg white protein was adsorbed probably due

to the higher molar mass of ovalbumin (44.5 kDa), compared with

b-lactoglobulin (18.5 KDa). Stewart and Swaigood (1993) used

molecular exclusion chromatography and proteins of known Stoke

radii and found that a-lactalbumin andb-lactoglobulin easily pen-

etrated the pores of alginate microparticles. In another recentstudy, the authors found that shorter polyelectrolyte chains

adsorbed larger quantities than larger chains (Malinova and

Wandrey, 2007).

The extents of protein adsorption onto films were 24.1 1.2%

and 25.3 1.0% for films coated with egg white protein (pH 3.5)

and whey protein (3.75), respectively, when 4% protein in solution

was used. The amounts of whey proteins or egg white proteins

adsorbed onto the films were similar to one another, however

the values were different from the results observed for the micro-

particles indicating that pore size could be an important factor for

the adsorption of small molecules. The amount of calcium bound to

the films was higher than in the microparticles indicating that

stronger and less porous gels were produced in the case of the

films. In a previous study in which films were produced by ionicgelation using pectin and alginate, a greater amount of calcium

and lower swelling were observed in the alginate films due to

the more effective cross-linking of calcium and carboxylic groups

when alginate was used (Sriamornsak and Kennedy, 2008).

The adsorption of proteins onto the microparticles was greater

than that onto the films, independent of the type of protein coating

used, at a concentration of 4% protein in solution. It is possible that

the smaller quantity of free carboxylic groups present after ionic

gelation resulted in a lower availability of carboxylic groups for

interaction with protein molecules in solution, therefore resulting

in lower quantities of adsorbed protein in the films. Additionally,

the smaller surface area available for electrostatic interactions

compared with that of the microparticles represents another char-

acteristic of the films that may be associated with the loweramount of adsorbed protein.

Fig. 2. Optical microscopy (columns 1 and 2) and scanning electron microscopy (column 3) images of the alginate/pectin particles produced by ionic gelation: (A) (pH 3.5)

and (D) (pH 3.75) without coating. (B and C) coated with OVA (pH 3.5) and (E and F) coated with WPC (pH 3.75) at a 4% protein concentration in solution. (A, B, D and E) OM

bar = 100 lm; (C and F) SEM bar = 10lm.

K.C. Aguilar et al. / Journal of Food Engineering 154 (2015) 1724 21

http://-/?-http://-/?-

7/25/2019 Aguilar 2015

6/8

Moisture contents of 96.2 0.1% and 96.3 0.0% were observed

for the films produced at pH 3.75 and 3.5, respectively. The mois-

ture contents of the protein-coated films (containing glycerol)

decreased to 93.6 0.4% and 93.8 0.0% when coated with whey

and egg white proteins, respectively, these values were signifi-

cantly different (p< 0.05) compared with those obtained for the

films without protein coatings.

3.6. Functional characterization of the alginate-pectin films

Because gel formation is very rapid, at least at the gel surface,

the formation of homogeneous films through casting and solvent

evaporation is difficult. Thus various alternatives have been tested

to improve the homogeneity of the gel films.Da Silva et al. (2009)

and Bierhalz et al. (2012) produced films by ionic gelation in two

steps. In the first step, small amounts of calcium were added to

the film-forming solution, which was subsequently homogenized

and dried. The films were then removed from the supports and

placed in a calcium solution to reinforce cross-linking and produce

stronger films. Using a different strategy, Sriamornsak and

Kennedy (2006)also used external ionic gelation, placing the poly-

saccharide solution over a dialysis membrane and placing the

membrane on the surface of a calcium solution. The membrane

allowed for the diffusion of calcium ions through the entire contactarea occupied by the film-forming solution; this method, according

to the authors produced strong homogeneous films. Another alter-

native for improving the homogeneity of the polysaccharide distri-

bution during the ionic gelation proposed by Draget et al. (1991)

involved the use of internal ionic gelation, with the addition of

an initially insoluble calcium salt to the film-forming solution. Sub-

sequently, through pH adjustment, the salt dissociated and became

available for association with carboxylic groups of the polysaccha-

ride present in the solution, thus initiating gelation and producing

a homogeneous distribution of the polysaccharide in the gel.

Recently, Benavides et al. (2012) used this alternative method toproduce alginate films containing oregano essential oil for anti-

microbial applications in foods.

In this study, an alternative strategy to produce such films

was tested. The polysaccharide film-forming solution was placed

on an acrylic support, and the support was placed in a closed

chamber. The calcium solution was then passed through sprayers

placed in the closed chamber, producing an environment satu-

rated with micro-droplets of calcium solution over the film-form-

ing solution. In preliminary tests, various calcium atomization

times, film-forming solution volumes and hardening times were

tested.

The results in Table 2 indicate that there was a significant

increase (p< 0.05) in the thickness of the coated films (FP +

OVA3.50 and FP + WPC3.75) relative to that of the uncoated films(FP3.50and FP3.75).

Fig. 3. SEM images of the alginate/pectin films obtained by SEM (left column, surface; right column, cross section). (A and B) (pH 3.75) without protein coating, (C and D)

(coated with WPC, pH 3.75) and (E and F) (coated with OVA, pH 3.5) at a 4% protein concentration in solution.

22 K.C. Aguilar et al./ Journal of Food Engineering 154 (2015) 1724

7/25/2019 Aguilar 2015

7/8

7/25/2019 Aguilar 2015

8/8

Fontes, L.C.B., Ramos, K.K., Sivi, T.C., Queiroz, F.P.C., 2011. Biodegradable edible films

from renewable sources-potential for their application in fried foods. Am. J.

Food Technol. 6, 555567.

Gbassi, G.K., Vandamme, T., Yolou, F.S., Marchioni, E., 2011. In vitro effects of pH,

bile salts and enzymes on the release and viability of encapsulated Lactobacillusplantarumstrains in a gastrointestinal tract model. Int. Dairy J. 21, 97102 .

Gombotz, W.R., Wee, S.F., 1998. Protein release from alginate matrices. Adv. Drug

Del. Rev. 31, 267285.

Grant, G.T., Morris, E.R., Ress, D.A., Smith, P.J.C., Thom, D., 1973. Biological

interactions between polysaccharides and divalent cations: the egg-box

model. FEBS Lett. 32, 195198.Harnsilawat, T., Pangsawatmanit, R., McClements, D.J., 2006. Influence of pH and

ionic strength on formation and stability of emulsions containing oil droplets

coated by beta-lactoglobulin-alginate interfaces. Biomacromolecules 7, 2052

2058.

Haynes, C.A., Norde, W., 1994. Globular proteins at solid liquid interfaces. Colloid

Surfaces B: Biointerfaces 2, 517566.

Hbrard, G., Hoffart, V., Beyssac, E., Cardot, J.M., Alric, M., Subirade, M., 2010. Coated

whey protein/alginate microparticles as oral controlled delivery systems for

probiotic yeast. J. Microencapsul. 27, 292302.

Humblet-Hua, K.N.P., Scheltens, G., Van der Linden, E., Sagis, L.M.C., 2011.

Encapsulation system based on ovalbumin fibrils and high methoxyl pectin.

Food Hydrocolloids 25, 307314.

Malinova, V., Wandrey, C., 2007. Loading polyelectrolytes onto porous

microspheres: impact of molecular and electrochemical parameters. J. Phys.

Chem. B 111, 84948501.

Martinsen, A., Storr, I., Skjrk-Brk, G., 1992. Alginate as immobilization material:

III. Diffusional properties. Biotech. Bioeng. 39, 186194.

Mestdagh, M.M., Axelos, M.A.V., 1998. Physico-chemical properties of

polycarboxylate gel phase and their incidence on the retention/release of

solutes. Biopolymer Sci.: Food Non Food Appl. 91, 303314 (Les Colloques 91,

INRA Ed.).

Oakenfull, D., Pearce, J., Burley, R.W., 1997. Protein gelation. In: Damodaran, S.,

Paraf, A. (Eds.), Food Proteins and Their Applications. Marcel Dekker, New York,

USA, pp. 111142.

Opanasopit, P., Apirakaramwong, A., Ngawhirunpat, T., Rajanarata, T., Ruktanonchai,

N., 2008. Development and characterization of pectinate micro/nanoparticles

for gene delivery. AAPSPharmSciTech 9, 6774.

Parris, N., Coffin, D.R., Remon, F.J., Pessen, H., 1995. Composition factors affecting

the water vapor permeability and tensile properties of hydrophilic films. J.

Agric. Food Chem. 43, 14321435.

Pranoto, Y., Rakshit, S.K., Salokhe, V.M., 2005. Enhancing antimicrobial activity of

chitosan films by incorporating garlic oil, potassium sorbate and nisin. LWT

Food Sci. Technol. 38, 859865.

Ralet, M.-C., Dronnet, V., Buchholt, H.C., Thibault, J.-F., 2001. Enzymatically and

chemically de-esterified lime pectins: characterisation, polyelectrolyte

behaviour and calcium binding properties. Carbohydr. Res. 336, 117125.

Ramos, O.L., Reinas, I., Silva, S.I., Fernandes, J.C., Cerqueira, M.A., Pereira, R.N., 2013.

Effect of whey protein purity and glycerol content upon physical properties of

edible films manufactured therefrom. Food Hydrocolloids 30, 110122.

Rhim, J.W., 2004. Physical and mechanical properties of water resistant sodium

alginate films. LWT Food Sci. Technol. 37, 323330.

Santipanichwong, R., Suphantharika, M., Weiss, J., McClements, D.J., 2008. Core

Shell biopolymers nanoparticles produced by electrostatic deposition of beet

pectin onto heat-Denatured b-lactoglobulin aggregates. J. Food Sci. 73, N23

N30.

Sezer, A.D., Akbuga, J., 1999. Release characteristics of chitosan treated alginate

beads: II. Sustained release of a low molecular drug from chitosan treatedalginate beads. J. Microencapsulation 16, 687696.

Smidsrod, O., 1974. Molecular basis for some physical properties of alginate in gels

state. Faraday Disc. Chem. Soc. 57, 263274.

Smrdel, P., Bogataj, M., Zega, A., Planinsek, O., Mrhar, A., 2008. Shape optimization

and characterization of polysaccharide beads prepared by ionotropic gelation. J.

Microencapsul. 5, 90105.

Soliman, E.A., El-Moghazy, A.Y., El-Din, M.S., Massoud, M.A., 2013.

Microencapsulation of essential oils within alginate: formulation and in vitroevaluation of antifungal activity. J. Enc. Adsorption Sci. 3, 4855.

Souza, F.N., Gebara, C., Ribeiro, M.C.E., Chaves, K.S., Gigante, M.L., Grosso, C.R.F.,

2012. Production and characterization of microparticles containing pectin and

whey protein. Food Res. Int. 49, 560566.

Sriamornsak, P., Kennedy, R.A., 2006. A novel gel formation method, microstructure

and mechanical properties of calcium polysaccharide gel films. Int. J. Pharm.

323, 7280.

Sriamornsak, P., Kennedy, R.A., 2008. Swelling and diffusion studies of calcium

polysaccharide gels intended for film coating. Int. J. Pharm. 358, 205213.

Sriamornsak, P., Kennedy, R.A., 2011. Effect of sodium fluorescein on release

characteristics of a macromolecule from calcium alginate beads. Carbohydr.

Polym. 84, 12081212.

Stewart, W.W., Swaigood, H.E., 1993. Characterization of calcium alginate pore

diameter by size exclusion chromatography using protein standards. Enzyme

Microb. Technol. 15, 922927.

Tello, F., Falfan-Corts, R.N., Martinez-Bustos, F., Da Silva, V.M., Hubinger, M.D.,

Grosso, C., 2015. Alginate and pectin-based particles coated with globular

proteins: production, characterization and anti-oxidative properties. Food

Hydrocolloids 43, 670678.

Tolstoguzov, V.B., 2003. Some thermodynamic considerations in food formulation.

Food Hydrocolloids 17, 123.

Ye, A., 2008. Complexation between milk proteins and polysaccharides via

electrostatic interaction: principles and applications a review. Int. J. Food

Sci. Technol. 43, 406415.

Zactiti, E.M., Kieckbusch, T.G., 2006. Potassium sorbate permeability in

biodegradable alginate films: effect of the antimicrobial agent concentration

and crosslinking degree. J. Food Eng. 77, 462467.

24 K.C. Aguilar et al./ Journal of Food Engineering 154 (2015) 1724

http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0085http://refhub.elsevier.com/S0260-8774(14)00549-4/h0085http://refhub.elsevier.com/S0260-8774(14)00549-4/h0085http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0100http://refhub.elsevier.com/S0260-8774(14)00549-4/h0100http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0110http://refhub.elsevier.com/S0260-8774(14)00549-4/h0110http://refhub.elsevier.com/S0260-8774(14)00549-4/h0110http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0120http://refhub.elsevier.com/S0260-8774(14)00549-4/h0120http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0130http://refhub.elsevier.com/S0260-8774(14)00549-4/h0130http://refhub.elsevier.com/S0260-8774(14)00549-4/h0130http://refhub.elsevier.com/S0260-8774(14)00549-4/h0135http://refhub.elsevier.com/S0260-8774(14)00549-4/h0135http://refhub.elsevier.com/S0260-8774(14)00549-4/h0135http://refhub.elsevier.com/S0260-8774(14)00549-4/h0140http://refhub.elsevier.com/S0260-8774(14)00549-4/h0140http://refhub.elsevier.com/S0260-8774(14)00549-4/h0140http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0150http://refhub.elsevier.com/S0260-8774(14)00549-4/h0150http://refhub.elsevier.com/S0260-8774(14)00549-4/h0150http://refhub.elsevier.com/S0260-8774(14)00549-4/h0155http://refhub.elsevier.com/S0260-8774(14)00549-4/h0155http://refhub.elsevier.com/S0260-8774(14)00549-4/h0155http://refhub.elsevier.com/S0260-8774(14)00549-4/h0160http://refhub.elsevier.com/S0260-8774(14)00549-4/h0160http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0170http://refhub.elsevier.com/S0260-8774(14)00549-4/h0170http://refhub.elsevier.com/S0260-8774(14)00549-4/h0170http://refhub.elsevier.com/S0260-8774(14)00549-4/h0175http://refhub.elsevier.com/S0260-8774(14)00549-4/h0175http://refhub.elsevier.com/S0260-8774(14)00549-4/h0180http://refhub.elsevier.com/S0260-8774(14)00549-4/h0180http://refhub.elsevier.com/S0260-8774(14)00549-4/h0180http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0190http://refhub.elsevier.com/S0260-8774(14)00549-4/h0190http://refhub.elsevier.com/S0260-8774(14)00549-4/h0190http://refhub.elsevier.com/S0260-8774(14)00549-4/h0195http://refhub.elsevier.com/S0260-8774(14)00549-4/h0195http://refhub.elsevier.com/S0260-8774(14)00549-4/h0195http://refhub.elsevier.com/S0260-8774(14)00549-4/h0200http://refhub.elsevier.com/S0260-8774(14)00549-4/h0200http://refhub.elsevier.com/S0260-8774(14)00549-4/h0205http://refhub.elsevier.com/S0260-8774(14)00549-4/h0205http://refhub.elsevier.com/S0260-8774(14)00549-4/h0205http://refhub.elsevier.com/S0260-8774(14)00549-4/h0210http://refhub.elsevier.com/S0260-8774(14)00549-4/h0210http://refhub.elsevier.com/S0260-8774(14)00549-4/h0210http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0220http://refhub.elsevier.com/S0260-8774(14)00549-4/h0220http://refhub.elsevier.com/S0260-8774(14)00549-4/h0220http://refhub.elsevier.com/S0260-8774(14)00549-4/h0225http://refhub.elsevier.com/S0260-8774(14)00549-4/h0225http://refhub.elsevier.com/S0260-8774(14)00549-4/h0225http://refhub.elsevier.com/S0260-8774(14)00549-4/h0230http://refhub.elsevier.com/S0260-8774(14)00549-4/h0230http://refhub.elsevier.com/S0260-8774(14)00549-4/h0230http://refhub.elsevier.com/S0260-8774(14)00549-4/h0230http://refhub.elsevier.com/S0260-8774(14)00549-4/h0230http://refhub.elsevier.com/S0260-8774(14)00549-4/h0230http://refhub.elsevier.com/S0260-8774(14)00549-4/h0225http://refhub.elsevier.com/S0260-8774(14)00549-4/h0225http://refhub.elsevier.com/S0260-8774(14)00549-4/h0225http://refhub.elsevier.com/S0260-8774(14)00549-4/h0220http://refhub.elsevier.com/S0260-8774(14)00549-4/h0220http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0215http://refhub.elsevier.com/S0260-8774(14)00549-4/h0210http://refhub.elsevier.com/S0260-8774(14)00549-4/h0210http://refhub.elsevier.com/S0260-8774(14)00549-4/h0210http://refhub.elsevier.com/S0260-8774(14)00549-4/h0205http://refhub.elsevier.com/S0260-8774(14)00549-4/h0205http://refhub.elsevier.com/S0260-8774(14)00549-4/h0205http://refhub.elsevier.com/S0260-8774(14)00549-4/h0200http://refhub.elsevier.com/S0260-8774(14)00549-4/h0200http://refhub.elsevier.com/S0260-8774(14)00549-4/h0195http://refhub.elsevier.com/S0260-8774(14)00549-4/h0195http://refhub.elsevier.com/S0260-8774(14)00549-4/h0195http://refhub.elsevier.com/S0260-8774(14)00549-4/h0190http://refhub.elsevier.com/S0260-8774(14)00549-4/h0190http://refhub.elsevier.com/S0260-8774(14)00549-4/h0190http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0185http://refhub.elsevier.com/S0260-8774(14)00549-4/h0180http://refhub.elsevier.com/S0260-8774(14)00549-4/h0180http://refhub.elsevier.com/S0260-8774(14)00549-4/h0180http://refhub.elsevier.com/S0260-8774(14)00549-4/h0175http://refhub.elsevier.com/S0260-8774(14)00549-4/h0175http://refhub.elsevier.com/S0260-8774(14)00549-4/h0170http://refhub.elsevier.com/S0260-8774(14)00549-4/h0170http://refhub.elsevier.com/S0260-8774(14)00549-4/h0170http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0165http://refhub.elsevier.com/S0260-8774(14)00549-4/h0160http://refhub.elsevier.com/S0260-8774(14)00549-4/h0160http://refhub.elsevier.com/S0260-8774(14)00549-4/h0155http://refhub.elsevier.com/S0260-8774(14)00549-4/h0155http://refhub.elsevier.com/S0260-8774(14)00549-4/h0155http://refhub.elsevier.com/S0260-8774(14)00549-4/h0150http://refhub.elsevier.com/S0260-8774(14)00549-4/h0150http://refhub.elsevier.com/S0260-8774(14)00549-4/h0150http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0145http://refhub.elsevier.com/S0260-8774(14)00549-4/h0140http://refhub.elsevier.com/S0260-8774(14)00549-4/h0140http://refhub.elsevier.com/S0260-8774(14)00549-4/h0140http://refhub.elsevier.com/S0260-8774(14)00549-4/h0135http://refhub.elsevier.com/S0260-8774(14)00549-4/h0135http://refhub.elsevier.com/S0260-8774(14)00549-4/h0135http://refhub.elsevier.com/S0260-8774(14)00549-4/h0130http://refhub.elsevier.com/S0260-8774(14)00549-4/h0130http://refhub.elsevier.com/S0260-8774(14)00549-4/h0130http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0125http://refhub.elsevier.com/S0260-8774(14)00549-4/h0120http://refhub.elsevier.com/S0260-8774(14)00549-4/h0120http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0115http://refhub.elsevier.com/S0260-8774(14)00549-4/h0110http://refhub.elsevier.com/S0260-8774(14)00549-4/h0110http://refhub.elsevier.com/S0260-8774(14)00549-4/h0110http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0105http://refhub.elsevier.com/S0260-8774(14)00549-4/h0100http://refhub.elsevier.com/S0260-8774(14)00549-4/h0100http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0095http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0090http://refhub.elsevier.com/S0260-8774(14)00549-4/h0085http://refhub.elsevier.com/S0260-8774(14)00549-4/h0085http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0080http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075http://refhub.elsevier.com/S0260-8774(14)00549-4/h0075