Journal of Food Engineering 118 (2013) 271278

Contents lists available at SciVerse ScienceDirect

Journal of Food Engineering

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

Effect of the incorporation of antioxidants on physicochemical and

antioxidant properties of wheat starchchitosan lms
J. Bonilla , E. Taln, L. Atars, M. Vargas, A. Chiralt
Instituto Universitario de Ingeniera de Alimentos para el Desarrollo, Universitat Politcnica de Valncia, Camino de Vera, s/n, 46022 Valencia, Spain

a r t i c l e

i n f o

Article history:
Received 24 January 2013
Received in revised form 3 April 2013
Accepted 13 April 2013
Available online 28 April 2013
Keywords:
Edible lms
Wheat starch
Chitosan
Basil essential oil
Thyme essential oil
a-Tocopherol
Citric acid
ABTS

a b s t r a c t
Blend edible lms were prepared from wheat starch (WS) and chitosan (CH) with glycerol as plasticizer.
Four active ingredients (antioxidants) were added, namely basil essential oil, thyme essential oil, citric
acid and a-tocopherol. The starch:antioxidant mass ratio was 1:0.1. Prior to characterisation, the lms
were conditioned at 25 C53%RH as to their structural, mechanical, optical and barrier properties. The
antioxidant capacity of the active ingredients was determined by means of a spectrophotometric method.
The incorporation of antioxidants led to a heterogeneous lm microstructure, mainly in those containing
a-tocopherol, which affected the surface roughness. Yellowness was induced in lms when a-tocopherol
was added and no notable colour changes were observed in the other cases, although all the antioxidants
increased the transparency of the lms. Despite of the fact that the mechanical properties were barely
affected by the incorporation of antioxidants, citric acid promoted an increase in the elastic modulus
but a decrease in lm stretchability. The water vapour barrier properties of the lms were only slightly
improved when citric acid and a-tocopherol were added, whereas the oxygen barrier properties were signicantly improved in all cases. The greatest antioxidant capacity of the lms was reported for lms containing a-tocopherol, which exhibited the highest antioxidant power.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
Edible lms or coatings are dened as thin layers of material
suitable for consumption, which act as a barrier against different
agents (water vapour, oxygen and moisture) thus increasing the
shelf-life of the product (Guilbert et al., 1996). Their use is a promising alternative in the eld of food preservation, allowing for the
reduction of waste plastics. Additionally, the incorporation of additives to these coatings, such as antimicrobials or antioxidant
agents, is potentially of interest (Baldwin et al., 1996).
Biodegradable polymers based on natural polysaccharides, particularly starch, can be produced at low cost and on a large scale.
The starch-based materials can reduce the use of non-renewable
resources. Furthermore, starch is one of the most abundant, natural
biopolymers. The starch granule is essentially composed of two
polysaccharides, amylose and amylopectin, and some minor components such as lipids and proteins. This polymer has attracted
considerable attention as a biodegradable thermoplastic polymer
(Angellier et al., 2006). The lms prepared from starch represent
promising applications in food packaging due to their biodegradability, low cost, exibility and transparency. However, several
authors have observed that, despite their ease of preparation,

Corresponding author. Tel.: +34 615766934; fax: +34 963877956.

E-mail address: jeannine_bonilla@hotmail.com (J. Bonilla).
0260-8774/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jfoodeng.2013.04.008

starch lms have some drawbacks, such as their poor mechanical

properties. For this reason, starch lms require the addition of
plasticizing compounds (Parra et al., 2004).
Chitosan is a cationic polysaccharide derived from the deacetylation of chitin, a component of the shells of crustaceans. It has
been shown to provide antimicrobial activity (Dutta et al., 2009;
Kumar, 2000), but its application is currently limited by its relatively high price and the fact that it is highly permeable to water
vapour (Butler et al., 1996). Bonilla et al. (2013) characterised lms
with varying proportions of wheat starch and chitosan, and observed that the lms with a chitosan proportion of 20% (w/w)
exhibited good antimicrobial properties and improved mechanical
properties as compared to pure wheat starch lms.
The addition of essential oils (basil and thyme) and other components with antioxidant activity, such as a-tocopherol (liposoluble antioxidant) or citric acid, can improve the functional
properties of edible lms and increase their potential use in the
preservation of foods with a high fat content. Despite the great potential of essential oils, their use in food preservation remains limited mainly due to their intense aroma, toxicity problems and
possible changes in the organoleptic properties of the food (Sanchez-Gonzalez et al., 2011a). The use of edible coatings to carry
essential oils could minimise the required doses by the encapsulation effect in the polymer matrix, which limits their volatilisation
and controls the compound release, thus reducing the negative impact of these ingredients.

272

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

Although the incorporation of bioactive compounds can modify

the barrier properties (water vapour and oxygen) of the lms,
doing so provides additional advantages, such as protection against
microbial growth and lipid oxidation (Gutirrez et al., 2008; Ahn
et al., 2008). In this sense, there is the challenge of nding appropriate formulations to delay the oxidation of lipids that are easily
oxidised in high-fat products (such as meat, sh or derivatives).
Recent studies have analysed the effect of the incorporation of
active compounds, such as antioxidants or antimicrobials, on physical and bioactive properties of different biopolymer lms. Sanchez-Gonzalez et al. (2011b) studied the effect of different
essential oils on properties of lms based on hydroxypropylmethylcellulose and chitosan. In this study, a great inuence of the polymer matrix on the changes induced by oil on mechanical, optical
and barrier properties of the lms was reported. Snchez-Gonzlez
et al. (2010a) developed chitosan lms containing tea tree essential
oil with antimicrobial properties against Penicilium Italicum and
Listeria monocytogenes, which showed lower mechanical resistance
and water vapour permeability than the oil-free lms, depending
on the oil:polymer ratio. The effect of basil and thyme essential oils
on properties of chitosan-based lms have been also analysed, as a
function of the homogenisation method of the lm forming dispersion (Bonilla et al., 2012). The oil ratio in the lm and the homogenisation conditions greatly affected mechanical, barrier and
optical properties of the lms. Chitosan and methylcellulose edible
lms with antioxidant properties were obtained by adding resveratrol to the polymer matrix (Pastor et al., 2013). Resveratrol improved the lm water barrier properties, but made the lms less
resistant and stretchable, more opaque and less glossy. Cinnamon
and ginger essential oils were incorporated into sodium caseinate
lms at low protein:oil ratio (less than 1:0.1) (Atars et al.,
2010). In these conditions, essential oils did not promote changes
in the lm mechanical properties, although slightly reduced water
vapour permeability. Despite the high antioxidant activity of the
oils, they did not improve the lm protection capacity against oxidation of sunower oil. The antioxidant activity of sodium caseinatecorn starch blend lms containing a-tocopherol was also
analysed by Jimenez et al. (2013). Ginger essential oil, citric acid
or ascorbic acid were incorporated into hydroxypropyl methylcellulose edible coatings to prevent lipid oxidation of roasted almonds
(Atars et al., 2011). A cross-linking effect in the lm matrices containing ascorbic or citric acid was detected through the analysis of
the lm microstructure, mechanical behaviour and oxygen and
water vapour barrier properties. These lms were the most effective protectors against oxidation of almonds.
The aim of this work was to study the effect of adding different
ingredients essential oils (basil and thyme), citric acid and atocopherol on the microstructure, mechanical behaviour, water
vapour and oxygen permeabilities and optical properties of wheat
starchchitosan lms. Additionally, the antioxidant properties of
the active compounds were quantied by means of a spectrophotometric method.

2. Materials and methods

2.1. Raw materials
The raw materials used were: native wheat starch (WS) with
21% amylose (Roquette Laisa SA Benifai, Valencia, Spain), high
molecular weight chitosan (CH) (Batch MKBB0585, SigmaAldrich,
Madrid, Spain), basil (B) and thyme (Th) essential oils (Herbes of
Moli, Alcoy, Alicante, Spain), a-tocopherol (Tp) and citric acid
(CA), (SigmaAldrich, Madrid, Spain), glycerol, acetic acid and
Mg(NO3)2 (Panreac Qumica SA, Castellar del Valls, Barcelona,
Spain).

The reactants for the antioxidant capacity assay Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid), K2S2O8 and
ABTS (20 2-azino-bis-[3-etilbenzotiazol-6-sulfonic acid]) were
supplied by SigmaAldrich (Madrid, Spain).
2.2. Experimental design and preparation of the lms
WS was diluted at 1% (w/w) in distilled water, and glycerol was
added in a starch:glycerol ratio of 1:0.15. This dispersion was
heated in a water bath at 95 C for 40 min to promote starch gelatinization. CH (1% w/w) was dispersed in an aqueous solution of
acetic acid (1% v/w) under magnetic stirring at 40 C and
450 rpm for 24 h. The two solutions were mixed in a mass ratio
WS:CH of 4:1. The active ingredients were added at 0.1 g/g of
starch. Hence, ve lm-forming dispersions (FFDs) were obtained,
namely WS:CH (control formulation), WS:CH:B (wheat
starch:chitosan:basil essential oil), WS:CH:Th (wheat starch:chitosan:thyme essential oil), WS:CH:CA (wheat starch:chitosan:citric
acid) and WS:CH:Tp (wheat starch:chitosan:a-tocopherol). All of
these were homogenised using rotorstator equipment (Yellow
Line DL 25 Basic, IKA, Janke y Kunkel, Germany) at 13,500 rpm
for 4 min and degassed at room temperature by means of a vacuum pump. In order to obtain the lms, the amount of FFD that
contain 1 g of solids were cast in framed and leveled polytetrauorethylene (PTFE) plates (150 mm of diameter), thus obtaining 1 g
of solids per plate in all cases. After drying for 48 h at 25 C and
50%RH, the lms were removed from the plates and conditioned
at 5 C58%RH or 25 C53%RH in cabinets containing saturated
solutions of Mg(NO3)2.
2.3. Characterisation of the lms
2.3.1. Film thickness
The lm thickness was measured using a Palmer digital
micrometer (Comecta, Barcelona, Spain) to the nearest 0.001 mm.
Four to six random positions were taken from each lm sample
and considered for the mechanical and barrier property assays.
2.3.2. Microstructure of the lms
The microstructural arrangement of the lm components was
studied by combining two techniques: scanning electron microscopy (SEM) and atomic force microscopy (AFM). SEM micrographs
of the surfaces and cross-sections of the lms were obtained using
a JEOL JSM-5410 electron microscope (Japan). To observe the sections of the lms, the specimens were cryofractured by immersion
in liquid nitrogen. The samples were mounted on copper stubs,
gold coated and observed using an accelerating voltage of 10 kV.
The surface of the lms was analysed by AFM using a Nanoscope III.a, Scanning Probe Microscope (Digital Instruments, Inc.
Santa Barbara, California). For each formulation, three samples
were cut (1  1 cm2). A three-dimensional image of a lm surface
area (50  50 lm2) was obtained in each test, and three images
were captured per formulation. Following the ASME B46.1 (1995)
method, the following statistical parameters describing the surface
roughness were calculated:
Average roughness (Ra): the arithmetic average of the absolute
values of the height deviations with respect to a mean surface (1).

Ra

1 XN
jZ j
J1 j
N

Rootmeansquare roughness (Rq): rootmeansquare average

of height deviations taken from the mean data plane (2).

s
PN 2
J1 Z J
Rq
N

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

Image Surface Area Difference (ISAD, %): relative difference

between the real and the geometric surface area.
2.3.3. Optical properties: translucency, colour and gloss
The surface reectance spectra of the lms were obtained using
a MINOLTA spectrocolorimeter, model CM-3600d (Minolta CO., Tokyo, Japan), illuminant D65 and observer 10. The measurements
were taken at three random points on the surface exposed to the
atmosphere during the drying step. The level of transparency/
opacity of the samples was determined through internal transmittance (Ti) calculated according to Eqs. (3)(5) (Hutchings, 1999),
where R0 is the reectance of the sample measured on a black
background, R is the reectance of the sample backed by a standard
white standard background, Rg is the reectance of the white background. Reectance measurements were performed between 400
and 700 nm.



1
R0  R Rg
R
2
R0  Rg

p
a2  1

q
2
T i a  R0 2  b

The reectance of an innitely thick layer of the material (R1)

was determined by means of Eq. (6), prior to obtaining the colour

coordinates: Lightness (L), Chroma (C ab ) (Eq. (7)) and hue (hab ) (Eq.
(8)). Moreover, the Whiteness index (WI) was calculated using Eq.
(9).

R1 a  b
C ab

q
2
a2 b

 
b

hab arctg 
a

8
2

WI 100  100  L 2 a2 b 1=2

Following ASTM D523 (1999), a glossmeter (Multi-Gloss 268,

Konica Minolta, Langenhagen, Germany) was employed to measure
the gloss at 20, 60 and 85 on two lms of each formulation, previously conditioned at 25 C53%RH. Six measurements were taken on the face exposed to the atmosphere during the drying
process.
2.3.4. Tensile properties
The mechanical properties of the lms were determined using a
Universal Testing Machine (Stable Micro Systems, TA.XT plus,
Haslemere, England) following the ASTM D882 (2001) standard
method. After conditioning at 25 C53%RH, 8 rectangular samples
(25 mm  100 mm) were cut per formulation, and the thickness of
each specimen was determined at four different positions. The
lms were mounted on lm-extension grips and stretched at
50 mm min1 until breakage. The curves of force versus distance
obtained in the test curves were transformed into stress (r) versus
Hencky strain (eH) curves. The parameters, EM (elastic modulus,
MPa), TS (tensile strength at break, MPa), and E, % (deformation
at the fracture point, expressed in percentage with respect to the
initial gage length), were obtained following ASTM standard method D882.
2.3.5. Moisture content, water vapour permeability and oxygen
permeability
Once equilibrated, three lm samples were taken per formulation to determine the moisture content by means of a gravimetric

273

method. Firstly, the specimens were desiccated in a vacuum oven

at 60 C for 24 h, and secondly, they were stored in cabinets containing P2O5 until they reached constant weight. The moisture content (MC) was expressed as the g of water per 100 g of the dried
lm.
The water vapour permeability of the lms was determined at
25 C and 53100 RH gradient. The WVP was determined through
a modication of the ASTM E96-95 gravimetric method for exible
lms (Mc Hugh et al., 1993). Three circular samples were cut
(3.5 cm in diameter) per formulation, selecting lms without visible defects, such as cracks or pores. The thickness of each sample
was determined at 6 random positions. Payne permeability cups
of 3.5 cm in diameter (Elcometer SPRL, Hermelle/s Argenteau, Belgium) were lled with 5 ml distilled water (RH = 100%). Saturated
solutions of magnesium nitrate were used to generate 53% RH inside the cabinets. The shiny side of the lms was exposed to the
atmosphere at the lowest RH (53% RH at 25 C). The cups were
weighed using an analytical balance (ME36S, Sartorius, Germany)
(0.00001 g) at intervals of 1.5 h for 24 h. The slope of the weight
loss versus time plot was divided by the exposed lm area to calculate the water vapour transmission rate (WVTR). The vapour
pressure on the lms inner surface was obtained from Eq. (10),
which corrects the effect of the concentration gradients established
in the head space of the cup.

WVTR

P  D  LnP  p2 =P  p1 
R  T  Dz

10

where P is the total pressure (atm); D is the diffusivity of water

through air at 25 C (m2/s); R is the gas law constant
(82.057  103 m3 atm kmol1 K1); T is the absolute temperature
(K); Dz is the mean height of stagnant air gap (m), considering
the initial and nal value; p1 is the water vapour pressure on the
solution surface (atm); p2 is the corrected water vapour pressure
on the lms inner surface in the cup (atm).
Finally, WVP was obtained using Eq. (11), where p3 is the water
vapour pressure on the outer side of the lm.

WVP

WVTR
 thickness
p2  p3

11

The oxygen permeability (OP) test was conducted by following

the ASTM Standard Method D3985-05 (2010). The lms were exposed to oxygen and nitrogen ows using the Ox-Tran equipment
(Model 1/50, Mocon, Minneapolis, USA) at 25 C53%RH. The tests
were performed in continuous mode. The OP was calculated by
dividing the oxygen transmission rate (cm3 mm m2 day1 atm1)
by the difference in oxygen partial pressure between the two sides
of the lm, and multiplying this by the average lm thickness. Two
replicates were made per formulation.
2.4. Antioxidant activity
The antioxidant capacity of the active ingredients was determined by means of a spectrophotometric method described by
Re et al. (1998). This method is based on comparing the antioxidant activity of the sample with that of a standard antioxidant,
Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid),
a vitamin E analogue.
ABTS (2,20 -azinobis(3-ethylbenzothiazoline-6-sulfonic acid)
diammonium salt) was dissolved in water at 7 mM and allowed
to react with a 2.45 mM potassium persulfate solution (nal concentrations) in the dark, for 16 h. During this period, ABTS radical
cation (ABTS+), a blue chromophore, was produced. The tests were
performed with ethanolic dilutions of the ABTS+ solution, whose
absorbance at 734 nm was 0.70 (0.02). The antioxidants were diluted in ethanol prior to the determination. Dilutions were made so
that the addition of a 10 ll-aliquot to 990 ll of ABTS+ dilution

274

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

would produce a decrease in absorbance of 2080% within 6 min.

Absorbance at 734 nm (A734) was registered every minute during
the test. For the calibration, Trolox standard solutions of up to
2 mM were prepared and the same procedure was followed. The
TEAC of the compounds was determined by comparing the corresponding percentage of absorbance reduction to the Trolox concentrationresponse curve and expressed as the mass of the
compound which produces the same percentage of absorbance
reduction as 1 mM Trolox solution. All the determinations were
carried out three times using a Beckman Coulter DU 730 (England)
spectrophotometer, with ethanol as the blank.
For the analysis in the lms, 0.1 g of each lm sample, cut into
small pieces, was previously extracted with 10 mL of a hydro-alcoholic mixture (1:1 volume ratio) to favour the extraction of the
antioxidants. To this end, samples were maintained under stirring
overnight. Afterwards, the extracts were submitted to the same
analytical procedure that was used for the active ingredients.
2.5. Statistical analysis
The statistical analysis of the data was performed through an
analysis of variance (ANOVA) using Statgraphics Plus (Manugistics
Corp., Rockville, MD). Fishers least signicant difference (LSD) procedure was used.
3. Results and discussion
3.1. Microstructure
The microstructure of the lms was qualitatively studied, with
the aim of observing the effect active ingredients have on the
arrangement of the lm components. Fig. 1 shows SEM micrographs of the lm surfaces in contact with air during drying and
Fig. 2 shows the cross section images of the cryofractured lms.
The lm without antioxidants (WS:CH) exhibited a compact and
ordered structure, without pores, which suggests that wheat starch

and chitosan are compatible polymers, able to form a homogeneous structure. The fact that the two polymers are highly compatible was previously observed by Bonilla et al. (2013). The addition
of active ingredients, which represent almost 7% of the total lm
solids, brought about some structural changes, as a function of
the nature of the added compound(s). Essential oils were integrated into the polymer matrix to differing degrees. The lm with
thyme essential oil (WS:CH:Th) did not show a marked phase separation, and it was only on the lm surface that coated lipid droplets were exhibited, resulting from the coalescence and creaming
of some droplets during the lm drying step; in the cross sections
of the lm, it was barely possible to observe very small lipid droplets. However, the lm with basil essential oil (WS:CH:B) exhibited
phase separation and the creaming of the oil was evident on the
lm surface, while oil droplets could be observed in the cross section of the polymer matrix. The incorporation of citric acid did not
lead to a highly heterogeneous starchchitosan lm (WS:CH:CA)
structure, although the lm surface and cross section appear coarser than for antioxidant-free lms. The greatest irregularities were
observed in the lms with a-tocopherol (WS:CH:Tp), due to its low
miscibility with the polymers and its different interactions with
each one. In fact, cross section micrographs of these lms showed
a two-layer separation: a smooth bottom layer, seeming to contain
a smaller amount of lipids, and a lipid-rich top layer. Likewise, big
droplets of one polymer phase containing a greater or smaller lipid
fraction can be observed in the other. This is especially clear at surface level and could indicate that polymers become more incompatible in the presence of a-tocopherol due to the different
interactions of this lipid with the polymer chains. A similar structural effect was observed for corn starchsodium caseinate blend
lms when a-tocopherol was added (Jimenez et al., 2013).
AFM images were obtained with the aim of quantifying the impact of active ingredients on the surface topography of the lms
(Fig. 3). The corresponding roughness parameters are shown in Table 1. The incorporation of the active compounds did not have a
relevant effect on the topography of the lm surface and, coherently, the roughness parameters were not affected with the only

WS:CH:B

WS:CH:Th

WS:CH:CA

WS:CH:Tp

Fig. 1. Scanning electron microscopy micrographs of the lm surface.

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

275

WS:CH

WS:CH:B

WS:CH:Th

WS:CH:CA

WS:CH:Tp

Fig. 2. Scanning electron microscopy micrographs of the cross-sections of the lms.

exception of a- tocopherol. WS:CH:Tp lms exhibited a rough surface, with an important increase in all roughness parameters
(p < 0.05), which coincides with the previously described phase
separation phenomena.
3.2. Optical properties
The lm gloss is related to the surface morphology (SnchezGonzlez et al., 2010b) and generally, the smoother their surface,
the glossier the lms are. The gloss values at 60 are shown in Table 1. This angle was selected following norm ASTM D523 (1999). A
signicant reduction in gloss (p < 0.05) was observed to take place
after the addition of every active ingredient, but mainly after those
which were lipid in nature. Essential oils and a-tocopherol provoked a reduction in the gloss of the WS:CH lms of over
50%,which is coherent with the less miscible nature of these compounds with the polymers and the subsequent effect of the dispersed particles on the lm surface. Although AFM did not detect
a signicantly higher degree of roughness in lms with essential
oils, this could be due to the low mechanical resistance of the lipid
droplets at the surface, which made it impossible to detect during
the surface scanning of the probe.
Table 2 shows the lightness, chroma and hue of the lms. The
incorporation of the antioxidants did not affect these variables very
markedly. However, the most intense yellow colour and the highest chroma were observed in lms containing a-tocopherol. As a
consequence, the whiteness index was the lowest in these lms.
Typical spectra of the internal transmittance of the lms between 400 and 700 nm are shown in Fig. 4. There were similarities

not only between the shape of the spectra of lms with antioxidants, but also between these and that of the WS:CH lm, which
is coherent with the slight colour differences observed for every
lm. However, it was found that all the antioxidants brought about
some increase in the internal transmittance values. This could be
explained by the formation of a less compact structure, with a lower refraction index, which, in turn, became less opaque. This is
coherent with the measured lm thickness (Table 3) which reveals
that thicker lms were obtained when antioxidants were present
in their formulation, despite the constant amount of solid per surface unit. This effect that the compact nature of the matrix had on
the lm transparency was partially mitigated in the case of the lm
containing a-tocopherol, due to the greater light dispersion provoked by the more heterogeneous lm structure.
3.3. Mechanical properties
Mechanical parameters, elastic modulus (EM), and tensile (TS)
and deformation (E, %) at break are shown in Table 3 for every lm
obtained. The mechanical behaviour of WS:CH lms is similar to
that reported by Bonilla et al. (2013) for lms of similar composition. The incorporation of active ingredients had variable effects on
the mechanical parameters, depending on their specic nature and
effects on the lm microstructure. Thyme essential oil did not signicantly affect the mechanical response of the lms, apart from a
slight decrease in stretchability. The lms containing basil essential oil or a-tocopherol showed some increase in EM and TS, as
compared to WS:CH lms, with no effect on the elongation at
break. On the other hand, the addition of citric acid led to an in-

276

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

WS:CH

WS:CH:Th

WS:CH:B

WS:CH:Tp

WS:CH:CA

Fig. 3. Typical AFM images of the surface topography of the lms.

Table 1
Average roughness (Ra), rootmeansquare roughness (Rq), relative difference
between the real and the geometric surface area (ISAD) and gloss at 60 of the lms.
Mean values and standard deviation in brackets.
Samples

Rq (nm)

WS:CH
WS:CH:B
WS:CH:Th
WS:CH:CA
WS:CH:Tp

147
109
212
109
486

(41)ab
(35)a
(42)b
(23)a
(125)c

Ra (nm)

ISAD (%)

Gloss (60)

113 (38)ab
85 (27)a
172 (37)b
86 (18)a
371 (124)c

0.8 (0.3)a
1.4 (0.2)ab
1.22 (0.07)ab
0.89 (0.03)a
4 (3)b

51
19
13
31
13

(16)c
(14)a
(2)a
(18)b
(3)a

Table 2

Lightness (L), hue (hab ), chroma (C ab ) and whiteness index (WI) values of the lm.
Average values and standard deviations in brackets.
Samples
WS:CH
WS:CH:B
WS:CH:Th
WS:CH:CA
WS:CH:Tp

L

hab
a

82 (2)
88 (3)b
84.4 (0.4)ab
86.9 (0.6)b
83.0 (0.4)a

96.6
98.2
95.2
97.8
89.5

(0.6)
(0.9)a
(0.2)a
(0.6)a
(0.6)a

C ab

WI

10.3 (0.9)a
11 (5)a
12.8 (0.6)ab
10.0 (0.4)a
15.7 (0.9)b

79 (2)ab
84 (6)b
79.8 (0.6)ab
83.5 (0.5)b
76.8 (0.9)a

The same superscript (abc) means homogeneous groups in LSD test.

The same superscript (abc) means homogeneous groups in LSD test.

crease in the elastic modulus and some reduction in lm resistance

and break and extensibility. This may be due to a cross-linking effect with the starch molecules, as previously described by Reddy
and Yang (2010).
3.4. Moisture content and barrier properties
The equilibrium moisture content of the lms stored at 53%RH
25 C is shown in Table 4, together with the water vapour permeability (WVP) and oxygen permeability (OP) values. Film composi-

tion was observed to have no notable effect on the equilibrium

moisture content, probably due to the relatively low level (6.5%
of total solids) of hydrophobic compound incorporated.
WVP was barely affected by the incorporation of active compounds. Although the incorporation of a-tocopherol or citric acid
led to some WVP reduction, essential oils had no signicant impact. In the case of citric acid, the above-mentioned cross-linking
effect may explain what can be observed, since a more closed matrix is formed, thus inhibiting the transfer of water molecules. The
polymer phase separation induced in the lms containing a-

277

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

0.90

Table 5
Trolox equivalent antioxidant capacity (TEAC) expressed as the amount (mg) of the
dry lm or compound which gives the same absorbance reduction as 1 mM Trolox
solution.

0.85

Ti

Antioxidant
WS:CH:B

0.80

WS:CH:Tp

WS:CH:CA

B
Th
CA
Tp

WS:CH

WS:CH:Th

0.75

0.70

400 420 440 460 480 500 520 540 560 580 600 620 640 660 680 700

(nm)
Fig. 4. Typical spectra of internal transmittance (Ti) of the lms.

Table 3
Tensile properties (elastic modulus: EM, tensile strength: TS and deformation: E %, at
break) and thickness of all lms equilibrated at 58% RH. Mean values and standard
deviations in brackets.
Samples

EM (MPa)

WS:CH
WS:CH:B
WS:CH:Th
WS:CH:CA
WS:CH:Tp

1458
1712
1450
1418
1598

(151)a
(46)b
(201)a
(605)a
(106)b

TS (MPa)

E (%)

44 (3)b
54.6 (1.2)c
43 (7)b
34 (12)a
47 (10)b

4.3
4.3
3.6
3.0
4.4

(0.8)b
(0,6)b
(0.6)a
(1.0)a
(1.4)b

Thickness (lm)
52.7 (1.6)a
57 (5)b
55 (11)b
52 (8)a
58 (5)b

The same superscript (abc) means homogeneous groups in LSD test.

Table 4
Moisture content (MC, g water/g dry lm) of lms equilibrated at 25 C53%RH.
Water vapor permeability (WVP, g mm kPa1 h1 m2) of the lms at 58100%RH
and 25 C and oxygen permeability (OP, cm3 mm2 atm1 day1) at 25 C53%RH.
Mean values and standard deviation in brackets.
Samples
WS:CH
WS:CH:B
WS:CH:Th
WS:CH:CA
WS:CH:Tp

WVP  1010
a

18 (2)
19.1 (1.4)a
18.2 (0.5)a
15.6 (1.0)b
15.6 (1.3)b

OP  102
6.7
3.7
3.2
3.8
3.8

(0.7)a
(1.0)b
(0.2)b
(0.7)b
(0.7)b

%MC
8.4 (0.5)a
8,67 (0.13)a
8.62 (0.16)a
8.8 (1.3)a
8.2 (0.3)a

The same superscript (abc) means homogeneous groups in LSD test.

tocopherol, which enhances the tortuosity index for mass transport

in the matrix (Perez-Gag and Krochta, 2001) may also explain the
observed decrease in the WVP values.
Oxygen permeability values at 25 C53%RH are also shown in
Table 4. The incorporation of antioxidants caused a signicant
reduction (of nearly 50%) of OP in every case, as compared to
WS:CH lms. The incorporation of liquid lipids, to hydrocolloid
lms usually implies an increase in the oxygen permeability due
to the enhancement of the oxygen solubility in the lm, which promotes the transfer of the oxygen molecules (Fabra et al., 2012).
Nevertheless, when the lipid shows antioxidant activity, this effect
is not observed. A chemical oxygen blocking effect produced by the
antioxidants could be responsible for this improvement in the oxygen barrier properties, as has been previously observed for cornstarch-sodium caseinate lms containing a-tocopherol, in contrast
with the negative effect produced by oleic acid in these lms (Jimenez et al., 2013).

3.5. Antioxidant activity

Table 5 shows the TEAC of the essential oils, citric acid and atocopherol together with that of the dry lms, which is dened

TEAC (mg)
Pure antioxidant

Dry lm

Antioxidant in dry lm

1.93
1.58
1.93
0.54

245
145
177
96

15.9
9.4
11.5
6.2

as the amount (mg) of sample that produces the same absorbance

reduction as 1 mM Trolox solution. The TEAC values of the antioxidants incorporated in the dried lms have been also estimated by
considering their mass fraction in the dry lms. Of all the pure
antioxidants, the lowest antioxidant capacity was found for basil
essential oil and citric acid, since the greatest amounts of these
are necessary to produce the same percentage of absorbance
reduction as 10 ll of 1 mM Trolox solution. Thyme essential oil
exhibited a slightly higher antioxidant power, and a-tocopherol
was the most effective antioxidant, as revealed by the test. As expected, lms showed lower antioxidant capacity than pure compounds since their mass fraction in the lm solids is relatively
low. Nevertheless, the observed trend in the antioxidant capacity
of pure compounds remained almost constant: the most antioxidant lm was that containing a-tocopherol, whereas the lms containing basil essential oil were the least effective. The antioxidant
capacity of the lms is thus related with that obtained for the
incorporated pure compounds, although their possible oxidation
during the lm formation and the interactions with the lm matrix
reduced the expected capacity. In fact, the TEAC values estimated
for the antioxidants in the lms were greatly increased with respect to those obtained for pure compounds (reduction of the antioxidant activity). This conrms the loss of the antioxidant capacity
of the compounds during lm formation and the extraction procedure, probably due to their partial oxidation. In the case of volatile
compounds, such as essential oils, the activity reduction may also
be related with their volatilisation during lm drying. In fact, the
estimated TEAC value for basil essential oil in the lm was higher
(lower antioxidant power) than that of citric acid, although no differences were observed when these pure antioxidants were tested,
which could be attributed to losses of the volatile compounds of
basil essential oil.
4. Conclusions
The incorporation of the antioxidants (basil and thyme essential
oils, citric acid and a-tocopherol) at 6.7% in wheat starch:chitosan
(4:1) lms gave rise to microstructural changes that were dependent on the different afnity of components. The greatest structural change was induced by a-tocopherol which promoted
polymer phase separation in the lm. All the antioxidants provoked a loss in lm gloss, but a gain in transparency, while atocopherol induced yellowness and a reduction of the whiteness
index. Although the mechanical properties of the lms were barely
affected by the antioxidants, citric acid promoted an increase in the
elastic modulus but a decrease in the tensile and deformation at
break, probably due to cross-linking with starch chains. Although
the water vapour barrier properties of the lms were only slightly
improved when citric acid and a-tocopherol were added, the oxygen barrier properties of the lms were signicantly improved in
every case, probably due to a chemical oxygen blocking effect.
The greatest antioxidant capacity of the lms was expected for
lms containing a-tocopherol, which showed the highest antioxidant power of all the tested compounds.

278

J. Bonilla et al. / Journal of Food Engineering 118 (2013) 271278

References
Ahn, J.H., Kim, Y.P., Seo, E.M., Choi, Y.K., Kim, H.S., 2008. Antioxidant effect of natural
plant extracts on the microencapsulated high oleic sunower oil. Journal of
Food Engineering 84, 327334.
Angellier, H., Molina Boisseau, S., Dole, P., Dufresne, A., 2006. Thermoplastic starch
waxy maize starch nanocrystals nanocomposites. Biomacromolecules 7, 531
539.
ASME B46.1, 1995. Surface Texture: Surface Roughness, Waviness and Lay, An
American National Standard, ASME, New York.
ASTM, 1995. Standard test methods for water vapor transmission of materials.
Standard Designations: E96-95. Annual Books of ASTM, ASTM, Philadelphia, pp.
406413.
ASTM, 1999. Standard test method for specular gloss. In Designation (D523). Annual
Book of ASTM Standards, vol. 06.01. American Society for Testing and Materials,
Philadelphia, PA.
ASTM, 2001. Standard test method for tensile properties of thin plastic sheeting. In
Standard D882. Annual Book of American Standard Testing Methods. American
Society for Testing and Materials, Philadelphia, PA, pp. 162170
ASTM, 2010. Standard test method for oxygen gas transmission rate through plastic
lm and sheeting using a coulometric sensor. Designations (3985-05).
Atars, L., Bonilla, J., Chiralt, A., 2010. Characterization of sodium caseinate-based
edible lms incorporated with cinnamon or ginger essential oils. Journal of Food
Engineering 100, 678687.
Atars, L., Prez-Masi, R., Chiralt, A., 2011. The role of some antioxidants in the
HPMC lm properties and lipid protection in coated toasted almonds. Journal of
Food Engineering 104, 649656.
Baldwin, E.A., Nisperos, M.O., Chen, X., Hagenmaier, R.D., 1996. Improving storage
life of cut apple and potato with edible coating. Postharvest Biology and
Technology 9, 15l163.
Bonilla, J., Atars, L., Vargas, M., Chiralt, A., 2012. Effect of essential oils and
homogenization conditions on properties of chitosan-based lms. Food
Hydrocolloids 26 (1), 916.
Bonilla, J., Atars, L., Vargas, M., Chiralt, A., 2013. Properties of wheat starch lmforming dispersions and lms as affected by chitosan addition. Journal of Food
Engineering 114 (3), 303312.
Butler, B.L., Vergano, P.J., Testin, R.F., Bunn, J.N., Wiles, J.N., 1996. Mechanical and
barrier properties of edible chitosan lms as affected by composition and
storage. Journal of Food Science 61, 953955.
Dutta, P.K., Tripathi, S., Mehrotra, G.K., Dutta, J., 2009. Perspectives for chitosan
based antimicrobial lms in food applications. Food Chemistry 114, 11731182.
Fabra, M.J., Talens, P., Gavara, R., Chiralt, A., 2012. Barrier properties of sodium
caseinate lms as affected by lipid composition and moisture content. Journal of
Food Engineering 109 (3), 372379.

Guilbert, S., Gontard, N., Gorris, L.G.M., 1996. Prolongation of shelf-life of perishable
food products using biodegradable lms and coatings. LebensmittelWissenschaft und-Technologie 29 (1,2), 1017.
Gutirrez, J., Barry-Ryan, C., Bourke, P., 2008. The antimicrobial efcacy of plant
essential oil combinations and interactions with food ingredients. International
Journal of Food Microbiology 124, 9197.
Hutchings, J.B., 1999. Food Colour and Appearance. Aspen Publishers, Maryland.
Jimenez, A., Fabra, M.J., Talens, P., Chiralt, A., 2013. Physical properties and
antioxidant capacity of starch-sodium caseinate lms containing lipids. Journal
of Food Engineering 116, 695702.
Kumar, M.N.V.R., 2000. A review of chitin and chitonan applications. Reactive and
Functional Polymers 46 (1), 127.
Mc Hugh, T.H., Avena-Bustillos, R., Krochta, J.M., 1993. Hydrophobic edible lms:
modied procedure for water vapor permeability and explanation of thickness
effects. Journal of Food Science 58 (4), 899903.
Parra, D.F., Tadini, C.C., Ponce, P., Lugo, A.B., 2004. Mechanical properties and water
vapor transmission in some blends of cassava starch edible lms. Carbohydrate
Polymers 58, 475481.
Pastor, C., Snchez-Gonzlez, L., Chiralt, A., Chfer, M., Gonzlez-Martnez, C., 2013.
Physical and antioxidant properties of chitosan and methylcellulose based lms
containing resveratrol. Food Hydrocolloids 30, 272280.
Perez-Gag, M.B., Krochta, J.M., 2001. Lipid particle size effect on water vapor
permeability and mechanical properties of whey protein/beeswax emulsion
lms. Journal of Agricultural and Food Chemistry 49 (2), 9961002.
Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., Rice-Evans, C., 1998.
Antioxidant activity applying an improved ABTS radical cation decoloration
assay. Free Radical Biology and Medicine 26, 12311237.
Reddy, Narendra, Yang, Yiqi, 2010. Citric acid cross-linking of starch lms. Food
Chemistry 118 (3), 702711.
Snchez-Gonzlez, L., Gonzlez-Martnez, C., Chiralt, A., Chfer, M., 2010a. Physical
and antimicrobial properties of chitosan-tea tree essential oil composite lms.
Journal of Food Engineering 98 (4), 443452.
Snchez-Gonzlez, L., Chfer, M., Chiralt, A., Gonzlez-Martnez, C., 2010b. Physical
properties of edible chitosan lms containing bergamot essential oil and their
inhibitory action on Penicillium italicum. Carbohydrate Polymers 82, 277283.
Sanchez-Gonzalez, L., Vargas, M., Gonzlez-Martnez, C., Chiralt, A., Chfer, M.,
2011a. Use of essential oils in bioactive edible coatings. Food Engineering
Reviews 3, 116.
Sanchez-Gonzalez, L., Gonzlez-Martnez, C., Chiralt, A., Chafer, M., 2011b. Effect of
essential oils on properties of lm forming emulsions and lms based on
hydroxypropylmethylcellulose and chitosan. Journal of Food Engineering 105
(2), 246253.