Packaging Res. 2017; 2:1–11
Research Article
Open Access
Angelique Mahieu*, Caroline Terrie, and Nathalie Leblanc
Role of ascorbic acid and iron in mechanical and
oxygen absorption properties of starch and
polycaprolactone multilayer film
DOI 10.1515/pacres-2017-0001
Received Nov 16, 2016; accepted Jan 08, 2017
Abbreviations
Abstract: A trilayer film based on thermoplastic starch
(TPS) for the core layer and poly(ϵ-caprolactone) (PCL) for
the skin layers was obtained by coextrusion. Ascorbic acid
and iron powder were added at respectively 15% and 1.5%
w/w in the TPS layer for their capacity to act as oxygen
scavenger, making the film usable as active food packaging. This study demonstrates that these compounds also
play a role in the interactions between the different layers. FTIR measurements show that ascorbic acid migrates
at the interface between TPS and PCL, where it acts as a
compatibiliser between both polymers, probably by creating new interactions between polar functions of both
polymers. This leads to a better adhesion of the different
layers, demonstrated by the increase of the adhesion energy from 4.10−3 N·mm−1 for the multilayer film TPS-PCL
to 12.10−3 N·mm−1 for the multilayer film containing the
active components. Thanks to this compatibilising effect,
the mechanical properties of the multilayer film containing ascorbic acid and iron are widely improved with an average maximal tensile strength of 7 MPa, against 3.7 MPa
for the multilayer film without the active components and
with an elongation at break of respectively 1450% against
290%. However, despite the hydrophobicity of PCL, the
water sorption of the TPS-based layer is only slightly reduced. The multilayer film shows active oxygen scavenging properties but the rate of this reaction is divided by two
compared to the active film without PCL layers (15 days to
reach less than 1% oxygen for the active film without PCL
layers and approximately 30 days to reach the same oxygen level with the multilayer active film).
AA
PCL
RH
TPS
ascorbic acid
polycaprolactone
relative humidity
thermoplastic starch
1 Introduction
A great number of studies are looking for a solution to the
environmental problem caused by waste and fossil fuels
dependence of traditional plastics. Biodegradable materials and/or made from renewable resources could be a solution. Starch is one of the most widely studied material
for the replacement of petroleum-based plastics thanks
to its natural abundance, its renewability and its ability to form thermoplastic films under destructuring and
plasticizing conditions [1, 2]. Thermoplastic starch (TPS)
shows a wide range of properties according to the plasticizer level and the forming process used [3]. Thermoplastic starch based materials could particularly be interesting for short life-time food packaging [4]. Moreover
active compounds such as antioxidant or antimicrobial
agents can be added in thermoplastic starch to obtain active packaging films [5]. Active packaging can be defined as
“a packaging in which subsidiary constituents have been
deliberately included in or on either the packaging material or the package headspace to enhance the performance of the package system” [6]. Oxidation is one of the
major causes of food degradation. So reducing the oxy-
Keywords: biodegradable active packaging film; multilayer; oxygen absorption; compatibility
*Corresponding Author: Angelique Mahieu: AGRI’TERR Research
Unit, UniLaSalle, 3 rue du Tronquet, CS 40118, 76134 Mont-SaintAignan Cedex, France; Email: [email protected]; Tel.:
+33 02 32 82 92 00; fax: +33 02 35 05 27 40
Caroline Terrie: AGRI’TERR Research Unit, UniLaSalle, 3 rue du
Tronquet, CS 40118, 76134 Mont-Saint-Aignan Cedex, France; Email:
[email protected]
Nathalie Leblanc: AGRI’TERR Research Unit, UniLaSalle, 3 rue du
Tronquet, CS 40118, 76134 Mont-Saint-Aignan Cedex, France; Email:
[email protected]
© 2017 A. Mahieu et al., published by De Gruyter Open.
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.
Unauthenticated
Download Date | 6/15/17 10:03 PM
2 | A. Mahieu et al.
gen rate in food environment allows to extend its shelflife. Moreover, O2 elimination inhibits aerobic microorganism development. Most of the oxygen scavenger packagings available in the market are developed from synthetic traditional polymers. Although lots of researches
concern biopolymers suitable for food packaging applications [7, 8], very few studies concern biomaterials for food
packaging which could have an action on food environment to extend its shelf-life. In a previous study [9] an
active biodegradable film based on thermoplastic starch
with active oxygen scavenger components (ascorbic acid
and iron powder) has been elaborated. This film can absorb all the oxygen present in the measuring chamber
within approximately 15 days. In this case, the high hydrophilicity of TPS film turns into an advantage because
the oxidation of the active components is activated by the
water absorbed. But some drawbacks of TPS water absorption remain: its gas permeability increases in high humidity environments [10] and its mechanical properties
highly depends upon humidity rate. Blending TPS with
hydrophobic and biodegradable polyesters such as polycaprolactone (PCL) [11, 12] or polylactic acid (PLA) [13]
is a widely proposed solution to improve these characteristics. PCL is also used with additives to develop antimicrobial and compostable food packaging [14]. It has
been shown, in previous work, that blending 30% PCL
(50 000 g/mol) with TPS leads to an homogeneous material [15]. The surface affinity for water of the material is reduced as well as the variability of their mechanical properties with humidity, even if the effect of the blend on film
water sorption is low. But when PCL is blended to TPS
matrix containing active components, the oxygen absorption kinetic decreases and the mechanical properties of the
film are not improved at high humidity rates [9]. OrtegaToro et al. have shown that TPS-PCL blend films present
poor mechanical properties due to the lack of compatibility between both polymers [16] but the same authors succeeded in reducing phase separation by citric acid incorporation [17]. Another alternative to associate PCL and TPS
can be in the form of a multilayer film [18]. PCL-TPS multilayer films with antimicrobial properties have already
been obtained [19]. The association of two polymers into
a multilayered structure can allow to improve the surface
properties of the film and combine the properties of each
polymer [20], leading to a film with reduced water sensitivity and good oxygen barrier properties, provided that both
polymer layers are compatible. The compatibility between
TPS and PCL can be improved by adding various molecules
such as methylenediphenyl diisocyanate [21], maleic anhydride [22], pyromellitic anhydride [23]. Nevertheless the
potential toxicity of these molecules may prevent the resulting materials from being used as food packaging.
The aim of this work is to realize an active film based
on starch, associated with polycaprolactone to improve resistance to water of the starch based layer. Contrary to the
previous work [9] and to improve the action of water protection of PCL on the starch based phase, in this work PCL
is added to the active starch-based matrix in the form of a
multilayer film by coextrusion. We will demonstrate that
ascorbic acid, besides its active property on oxygen, can
act as a compatibiliser between TPS and PCL. The properties of the multilayer films obtained are characterized
to investigate the effect of the PCL layers on the mechanical resistance and the permeability of these films. Thanks
to its ability to reduce the oxygen rate in the packaging
headspace and its need to be activated by water absorption, this active film could be envisaged to package short
life-time food which is sensible to oxidation and which
emits water vapor, like fruits and vegetables.
2 Experimental
2.1 Materials
Unmodified wheat starch, ascorbic acid, iron powder
and glycerol were obtained from Sigma Aldrich (France).
Wheat starch contains 26% amylose and 74% amylopectin.
The iron powder purity is more than 99% and its diameter
is 100 +/−20 µm. Poly(ϵ-caprolactone) (PCL CAPA 6500)
was purchased from Perstorp (UK). Its average molecular
weight is 50 000 g·mol−1 .
2.2 Sample preparation
In a first step starch was blended with 20% glycerol in
a thermo-regulated turbo-mixer and heated in an oven
at 170∘ C for 1 hour, allowing diffusion of glycerol into
the starch granules. 20% glycerol has been shown to be
the maximum rate leading to starch plasticization withTable 1: formulations of the starch-based layer (% w/w)
Formulation
(%w/w )
TPS
TPS - Fe
TPS - AA
TPS-AA-Fe
Wheat
Starch
80
78.8
68
66.8
Glycerol
20
19.7
17
16.7
Iron
powder
/
1.5
/
1.5
Ascorbic
acid
/
/
15
15
Unauthenticated
Download Date | 6/15/17 10:03 PM
Ascorbic acid and iron in mechanical and oxygen absorption properties
out phase separation between starch and glycerol [24, 25].
Then according to the various formulations (Table 1), active compounds were added in the starch-glycerol blend
and the humidity rate was adjusted at 10% w/w. The
amount of active compounds used is chosen to allow
the absorption of all the oxygen present in the experimental device (representing the volume of a standard
packaging) [9]. The resulting mixture was extruded and
granulated in a single screw extruder (SCAMEX, France)
equipped with two heating zones regulated at 100∘ C, then
110∘ C around the screw and 90∘ C for the die, at a screw
speed of 50 rpm.
In a second step, these compounds were extruded
through a coextrusion slit die together with PCL coming
from a second extruder. Three-layer films were obtained,
with TPS containing active-compounds as the inner layer
and PCL as the outer layers. The thicknesses of the layers
were modulated through regulation of the extruders screw
speeds, although they could not be accurately controlled.
Screw speeds of both extruders were adjusted to obtain
multilayer films with approximately 30% (w/w) of PCL and
70% of starch-based layer. Regardless of the settings, the
3-layer films always exhibited an upper layer much thicker
than the lower layer (up to 10 times). The path in the coextrusion die, causing the polymer to divide into two parts to
form the upper and lower layers, associated with the high
fluidity of the PCL at the processing temperatures, lead to
an easier access of the PCL towards the upper layer. After
extrusion films were cooled by calendering. Some multilayer active films were perforated across their upper PCL
layer with a 150 µm diameter needle. The perforation density was 8 holes per cm2 . All films were stored in controlled
chambers at 53% RH.
Various parameters are known to influence the properties of thermoplastic starch, in particular the nature
and amount of plasticizer [26], the humidity rate in the
film [15], the treatment used to plasticize starch [27] and
the material ageing [26]. All the materials studied here
contain 20% w/w of glycerol relative to the proportion of
starch, they have been extruded on the same equipment
with the same parameters and have been conditioned at
the same humidity rate during two weeks before testing.
So we may consider that the observed differences in the
film behaviors are due to the modification of the formulation and to the coextrusion with PCL.
| 3
2.3 Fourier transform infrared spectroscopy
(FTIR)
The FTIR spectra were recorded on a Nicolet iS10 spectrophotometer (Thermo scientific) equipped with a diamond crystal in attenuated total-reflectance mode (ATR). A
series of 64 scans were collected for each measure over the
4000 to 650 cm−1 spectral range with a 4 cm−1 resolution.
Spectra were analyzed with the Omnic 8.1 software. Measurements are performed separately on the film surface
or on the film core. The “core” measurement was realized
at 25% of the thickness below the surface. TPS spectrum
presents a characteristic peak at 938 cm−1 corresponding
to C-O-C bond of starch [28], which is not present in AA
spectrum. The spectrum of ascorbic acid presents a characteristic peak at 1750 cm−1 corresponding to the carbonyl
group and which is not present in TPS. Similarly to the PCL
enrichment factor previously used to study TPS-PCL blend
films [15], an ascorbic acid enrichment factor can be defined according to the equation 1:
AA surface enrichment factor by
(︁
)︁
A AApeak
A TPSpeak
FTIR = (︁
A AApeak
A TPSpeak
(1)
)︁surface
core
where A is the area of the characteristic peaks of TPS or
ascorbic acid.
Series of 6 experiments were carried out on each sample to calculate an average AA surface enrichment factor
and the corresponding standard deviation.
2.4 Water sorption measurements
Water vapor sorption isotherms of each sample were deduced from the static-gravimetric method by using saturated salt solutions at 23∘ C. These solutions were prepared
to generate an environment with controlled humidity in a
closed chamber. Circular samples of 5 cm in diameter of
each material were cut. The samples were covered with
hot-melt adhesive on the edges (which are not covered by
PCL), thus ensuring that the measurement is only representative of the water vapor which is absorbed by the films
through the PCL layers. The samples were conditioned at
a relative humidity close to ambient (53% HR) before testing then were placed in chambers with different water
activities obtained with above-saturated salt solutions of
Mg(NO3 )2 (0.53), CoCl2 (0.65), NaCl (0.75), KCl (0.85) and
KNO3 (0.94) respectively [29]. Samples were weighted until
equilibrium for 1 month to determine their maximal water
Unauthenticated
Download Date | 6/15/17 10:03 PM
4 | A. Mahieu et al.
content. The results presented are an average of three measurements on each studied film. Water sorption isotherms
were obtained by plotting the average maximal water content of the material (with the corresponding standard deviations) as a function of water activity.
and the permeability in cm3 · µm/m2 /24h/bar of each sample is calculated as a function of the thickness of the samples. The results are averages of three measurements on
each sample.
2.7 Peel test
2.5 Oxygen absorption
To measure their oxygen absorption capacity, samples
(constant weight of 5 grams) were placed in hermetic measuring chambers (glass bottles of 250 mL) whose corks
are equipped with a high gas barrier septum. As for water sorption measurements, the sample edges were covered with hot melt adhesive to measure only the oxygen
absorption due to the oxygen which go through the PCL
layers. The relative humidity in the bottles was adjusted at
94% RH with a saturated salt solution of KNO3 (constant
volume of 20 mL). Indeed it has been demonstrated previously [9] that the oxygen absorption rate increases with the
relative humidity in the measuring chamber because the
oxidation of ascorbic acid and iron needs to be activated by
water absorbed by the starch-based film. The bottles contain 225 cc of ambient air (20,9% O2 ). The samples were
fixed in the bottles so as to avoid any contact with the salt
solution. As in the method used by Buyn et al. [30], the evolution of the oxygen content in the bottle headspaces was
measured as a function of time thanks to a gas analyzer
(Oxybabyr 6.0). A sampling needle with a 0.45 µm PTFE
filter was inserted and 5 cc headspace gases were sampled
through the bottle septum. Calibration of the gas analyzer
was realized every day with pure nitrogen, pure carbon
dioxide and ambient air. Each measurement was carried
out in triplicate at room temperature. Average results of
these 3 measurements and corresponding standard deviations are presented in the result section.
The adhesion strength between the TPS matrix and the
PCL layers was evaluated by a T-peel test using a universal testing machine (TA.XTplus texture analyzer). Samples
of 200 mm long and 15 mm wide were cut in the centre of
the films and in the extrusion drawn direction and conditioned at 53% RH. Before the test about 50 mm long of PCL
layer was hand peeled in order to fix the core layer in a grip
and the PCL layer in the other grip. The peel tests were performed at ambient temperature and humidity with a separation rate of 60 mm·min−1 and an angle of 180∘ between
both layers. The force (in N) necessary to delaminate the
film is measured as a function of the distance (in mm). The
adhesion energy, expressed in N·mm−1 , was determined
from the average load (in steady-state conditions) and the
width of multilayer films according to Equation 2. The test
was repeated on 8 samples for each film to determine an
average adhesion energy and the corresponding standard
deviation.
Gp =
F
b(1 − cos θ)
(2)
Where Gp is the adhesion energy in N·mm−1 , F is the average force in N, b is the width of sample in mm and θ is the
angle between both layers
So
Gp =
2F
b
(3)
for the T-peel test.
2.6 Oxygen permeability
2.8 Tensile properties
Oxygen permeability of the multilayer films was measured
with a Mocon Ox-Tran 2/61, USA. The film was placed on
an aluminum mask with an open testing area of 5 cm2 . The
test cell was composed of two chambers separated by the
film. The water activity of the two gases was controlled by a
humidifier. Oxygen was introduced in the upstream compartment of the test cell. Oxygen transferred through the
film was conducted by the carrier N2 /H2 gas to the coulometric sensor. The tests were carried out at 23∘ C and at two
different relative humidities: 53% and 75% RH. The oxygen permeation rate (OTR) in cm3STP /m2 /24h was measured
Tensile tests were carried out by using a universal testing
machine (TA.XTplus texture analyzer, France), equipped
with a load cell of 500 N and following the international
standard ASTMD882. Average values and standard deviations of ten different standard dumbbell specimens were
reported.
The samples have been taken in the centre of the extruded films and in the extrusion drawn direction. The
samples geometry is a standard traction specimen [31].
The thickness of each sample was determined at four different positions. The films were placed in the grips and
Unauthenticated
Download Date | 6/15/17 10:03 PM
Ascorbic acid and iron in mechanical and oxygen absorption properties
| 5
Table 2: Mechanical characteristics of the studied films (average value ± standard deviation).
Formulation
TPS
TPS-AA
TPS-AA-Fe
PCL/TPS-AA/PCL
PCL/TPS-AA-Fe/PCL
PCL
Maximal tensile strength,
σm (MPa)
3.8 ± 0.3
0.46 ± 0.08
0.65 ± 0.06
3.7 ± 0.3
7.0 ± 1,4
46 ± 4
(a)
Modulus of Elasticity, E
(Mpa)
35 ± 10
5±2
6±2
29 ± 3
18 ± 2
167 ± 20
Elongation at break (%)
130 ± 20
160 ± 25
120 ± 12
290 ± 25
1450 ± 140
1500 ± 90
(b)
Figure 1: Stress-strain curves of the different films containing TPS (a) and the pure PCL (b).
stretched at 5 mm·min−1 until breakage. Prior to analysis,
the samples were conditioned under controlled humidity
(53% RH), and the tests were performed at room temperature and humidity, under atmospheric pressure. The average parameters, E (elastic modulus, MPa), σ m (tensile
strength at break, MPa) and ϵ (deformation at the fracture
point, expressed in percentage) were calculated.
3 Results and discussion
3.1 Mechanical properties
The mechanical resistance of the multilayer films was
characterized by tensile tests. The average stress-strain
curves of the different formulations are presented on Figure 1. Table 2 includes the mechanical characteristic values, maximal tensile strength, modulus of elasticity and
elongation at break for the different tested films. In the
experimental conditions, the pure thermoplastic starch
presents a classical plastic behavior. As explained previously [9], the addition of 15% (w/w) ascorbic acid leads to
a strong decrease of the maximal tensile strength which
is less than 0.5 MPa for TPS-AA. The effect of the addition of 1.5% (w/w) iron on the mechanical properties
is weakly significant. The elongation at break decreases
from 160% for TPS-AA to 120% for TPS-AA-Fe and the
maximal strength is only slightly improved. The mechanical behavior of the multilayer films is very different. The
PCL/TPS/PCL film could not be tested because the PCL layers delaminated very easily (the layers began to separate
from the cut of the tensile specimens). On the contrary
tensile specimens of PCL/TPS-AA/PCL and PCL/TPS-AAFe/PCL could be cut without delaminating. The mechanical behavior of these multilayer films is largely improved
thanks to the PCL layer in comparison with TPS-AA and
TPS-AA-Fe films.
Figure 1 (b) shows that the pure PCL film presents a
viscoelastic behavior with a high resistance and a high deformation capacity. A yield point appears at approximately
15 MPa and 12% elongation, followed by a necking step
where elongation goes on at a constant strength, and finally strength and elongation increase until the breaking
occurs at 45 MPa and 1500% elongation. The average ten-
Unauthenticated
Download Date | 6/15/17 10:03 PM
6 | A. Mahieu et al.
Table 3: Thickness of each layer (upper layer : PCLu , lower layer :
PCLl ), peeling force and corresponding adhesion energy for each
sample (average value ± standard deviation).
Figure 2: An example of a peeling curve obtained for the PCL/TPSAA-Fe/PCL film.
sile curves of the multilayer films PCL/TPS-AA/PCL and
PCL/TPS-AA-Fe/PCL follow a similar pattern. The yield
points of these both films occur at strengths close to the
maximal strength of TPS, respectively 3.3 and 3.9 MPa. After a little necking step, the elongation continues to increase, only until 290% for the multilayer film with TPSAA but until 1450% for the multilayer film with TPS-AA-Fe,
which is close to the elongation of pure PCL. The maximal strength for the multilayer film PCL/TPS-AA-Fe/PCL is
approximately 7.0 MPa, that is to say more than ten times
higher than without PCL. The improvement of the mechanical properties of the film TPS-AA-Fe by coextrusion
with PCL is thus very high, although for some specimens
a breakage of the TPS based layer was observed before the
breakage of the PCL outer layers. This can be explained by
a lack of homogeneity lengthways of the multilayer film,
since the TPS granules are supplied manually in the laboratory scale extruder. This may induces variability in TPS
flow, whereas the PCL flow is more constant. An extruder
with a more easily achieved steady-state would give more
homogeneous multilayer films, leading to better repeatability in mechanical properties. Nevertheless, PCL/TPSAA-Fe/PCL presents a good average mechanical resistance,
twice that of pure TPS film and an average elongation at
break almost equal to that of pure PCL film. These properties suggest that ascorbic acid associated to iron plays a
role of compatibiliser between the TPS-based and the PCL
layers.
The adhesion between the different layers of the multilayer films was measured using a T-peel test. As previously explained, the three-layer films exhibit differences in
thickness between the lower and the upper layers. The results of the peel test are then separated between the thicker
and the thinner layers (corresponding respectively to the
upper and the lower layers). An example of the curves
obtained by this method is presented on the Fig. 2. The
steady-state is rapidly reached. Some heterogeneity on the
Formulation
Thickness
(mm)
Force (N)
PCLu /
TPS/
PCLl
PCLu /
TPS-Fe/
PCLl
PCLu /
TPS-AA/
PCLl
PCLu /
TPS-AA-Fe/
PCLl
0.22 ± 0.05
0.55 ± 0.06
0.02 ± 0.01
0.27 ± 0.01
0.7 ± 0.1
0.04 ±0.03
0.30 ± 0.08
0.9 ± 0.2
0.09 ± 0.01
0.30 ± 0.03
0.8 ± 0,1
0.05 ± 0.01
0.03 ± 0.01
Adhesion
energy,
Gp (N·mm−1 )
0.4.10−3
0.3 ± 0.1
0.06 ± 0.04
4.10−3
0.8.10−3
0.3 ± 0.2
0.17 ± 0.06
3.7.10−3
2.2.10−3
0.5 ± 0.1
0.26± 0,06
7.1.10−3
3.4.10−3
0.9 ± 0.2
12.10−3
PCL layer can create the peaks observed on the curve. The
average thicknesses of each layer and the average peeling
forces required to separate the PCL layers from the TPSbased layer of each formulation are given in Table 3, as
well as the corresponding adhesion energy. In all cases the
thicker layer presents a lower adhesion force than the thinner layer. It is clear that a thickness of the PCL layer of less
than 100 µm is necessary to have a good adhesion with
the TPS-based layer. The maximum peeling force obtained
is 0.9 N for the PCL/TPS-AA-Fe/PCL. It is not a very high
value but it corresponds to a sufficient adhesion between
the layers to have a film with a good cohesion. Couturaud
et al. [32] have obtained peeling strength of the same order of magnitude on bilayer films of cellulose and PLA.
The formulation of the starch-based layer plays an important role on the adhesion with the PCL layer. We can see
on Table 3 that addition of iron powder in the TPS matrix
does not have any effect on the adhesion between layers.
On the contrary with addition of ascorbic acid in the TPS
matrix, the peeling force increases from 0.3 to 0.5 N. This
confirms the role of compatibiliser of the ascorbic acid between plasticized starch and PCL. The increase of peeling
force is even more important when ascorbic acid and iron
powder are associated in the TPS matrix (0.9 N for the thinner layer). These results confirm that the complex formed
by the ascorbic acid with the iron particles is a better compatibiliser than the ascorbic acid alone.
Comparable results have been obtained in a recent
study, Ortega-Toro et al. [33] have shown that ascorbic acid
sprayed at the interface between layers of TPS and PCL
Unauthenticated
Download Date | 6/15/17 10:03 PM
Ascorbic acid and iron in mechanical and oxygen absorption properties
| 7
Figure 3: FTIR spectra of TPS, pure AA and TPS-AA-Fe film at the surface and in the core.
can favor the layer adhesion. In our work this effect is all
the more important when ascorbic acid is associated with
iron particles. In the presence of iron, ascorbic acid can
form a ferrous ascorbate which contributes to the catalytic
effect of iron on the ascorbic acid oxidation [34]. The ferrous ascorbate seems to be very soluble in TPS matrix. It
could migrate at the interface between TPS and PCL layers
and create new interactions between polar groups of both
polymers, thus improving their compatibility. The compatibiliser effect of ascorbic acid can also be attributed to its
plasticizing role on the TPS matrix. As shown previously,
addition of ascorbic acid on TPS formulation leads to a
decrease of the film strength which can be attributed to a
plasticizing effect of the ascorbic acid on the TPS matrix.
This effect is associated with a decrease of the TPS viscosity in the extruder. The TPS viscosity is then closer to that
of PCL, which can lead to a better compatibility between
the both polymers [12].
3.2 FTIR analysis of ascorbic acid migration
To highlight the migration of ascorbic acid at the surface
of the film TPS-AA-Fe, FTIR analysis was carried out at the
surface and on the core of the film. The obtained spectra
are presented on Fig. 3 and compared with the FTIR spectra of the surface of TPS film and of pure ascorbic acid. TPS
and AA display very different FTIR spectra. The FTIR spectrum of TPS-AA-Fe at the surface of the film is very simi-
lar to AA spectrum, indicating that AA is present in large
quantity at the film surface. On the contrary, the FTIR spectrum of TPS-AA-Fe taken at a quarter of film thickness below the surface is very similar to TPS spectrum, indicating
that very few ascorbic acid is present at this position in the
film. The AA surface enrichment factor between surface
and core of the film TPS-AA-Fe is about 20 which means
that there is significantly more ascorbic acid towards the
surface of the film than in the core. So during extrusion of
TPS-AA-Fe film, ascorbic acid or the complex formed with
iron migrates towards the film surface. In the case of multilayer film ascorbic acid is thus located mostly at the interface between TPS-based and PCL layers, where it can act
as a compatibiliser.
3.3 Water vapor absorption
The mass of water absorbed by the different films as a function of the relative humidity is reported in Fig. 4 (The uncertainty of these measurements were calculated but they
are very low and the corresponding error bars are smaller
than the symbols of the points). The sorption test starts
at 53% RH because the samples were conditioned at this
humidity, and were not dried before testing to avoid any
layer delamination. So we compare the mass of water absorbed between the different films starting from the ambient humidity rate. As shown on Fig. 4, pure PCL does not
absorb any water due to the hydrophobic nature of PCL. On
Unauthenticated
Download Date | 6/15/17 10:03 PM
8 | A. Mahieu et al.
the contrary the hydrophilic nature of TPS is clearly visible
since the TPS film absorbs up to 24% (w/w) of water at 94%
RH. The multilayer films absorb a little less water than the
TPS film but the decrease is lower than that we could have
expected since the PCL does not absorb water and covered
entirely the TPS layer. PCL layers represent 30% of the multilayer film weight. Pure PCL does not absorb water. As we
can see on Fig. 4 the multilayer films based on TPS, TPSFe and TPS-AA absorb approximately 30% less than the
TPS film. This decrease of water absorbed can thus be attributed to the diminution of the proportion of hydrophilic
starch based matrix in the films and not to a protective effect of the PCL layers on the TPS-based layer. The mass of
water absorbed by the film PCL/TPS-AA-Fe/PCL is approximately 40% less than the TPS film at 94% RH, which corresponds to a slight decrease of water absorbed by the TPSbased layer. This last result is due to the better adhesion
between TPS containing ascorbic acid and iron and PCL
layers. But this decrease remains low: coextrusion with
PCL does not efficiently protect TPS-based layers from water absorption. Ortega-Toro et al. [33] have measured a very
little water vapor permeability of a PCL film. Their bilayer
films of TPS-PCL have low water-vapor permeabilities too,
but they were obtained by compression molding and not
by coextrusion. In our study, problems of heterogeneity of
the PCL layers could probably explain the obtained results.
The PCL layers do not prevent TPS-based layer from water
absorption but they reduce the absorption speed. Reaching the equilibrium mass in the humidity controlled chambers takes more time for the multilayer films than for the
pure TPS at high humidity rates. Between 53% and 75% HR
the equilibrium is reached in approximately 10 days for all
the films. At 85% and 94% HR the TPS film reaches equilibrium in approximately 15 days whereas the multilayer
films need more than 22 days to reach this point.
3.4 Oxygen permeability and absorption
capacity of the multilayer films
As shown on Table 4 the oxygen permeability of PCL film is
very high at both tested relative humidities whereas TPSAA-Fe film presents good barrier properties to oxygen at
low water content. The oxygen permeability of the TPS-AAFe film slightly increases with relative humidity (so with
water absorption of the film) but the value obtained at 75%
RH is still low. Dole et al. [35] have also observed that the
oxygen permeability of starch based films increases with
the increase of plasticizer content (water included). Oxygen permeability measured for PCL/TPS-AA-Fe/PCL also
increases with relative humidity. This behavior confirms
Figure 4: Moisture sorption curves at 23∘ C for TPS film, PCL and the
multilayer films.
Table 4: Oxygen permeability (OP) of different films, measured at 53% and 75% RH and 23∘ C. (Detection limit (D.L.):
0.1 cm3 /m2 /24 h)
films
TPS-AA-Fe
PCL
PCL/TPS-AA-Fe/PCL
OP (cm3 · µm/m2 /24h/bar)
53% RH
75% RH
< D.L.
92 ± 80
55477 ± 87
56424 ± 165
514 ± 78
802 ± 84
that PCL layers did not efficiently preserved TPS-based
layer from water. Indeed, if TPS based layer did not absorb
water, the oxygen barrier properties of the multilayer film
would have remained stable in spite of the relative humidity increase.
Nevertheless, the results show rather low values of
oxygen permeability for the multilayer film PCL/TPS-AAFe/PCL. These values are intermediate between these of
pure PCL and TPS-AA-Fe. If we consider that the oxygen
barrier effect of the multilayer film is only due to the TPSbased layer, the permeability value measured can be reported to the proportion of the TPS-based layer, that is to
say to 70% of the multilayer film total thickness. In that
case, we obtain OP values of 358 cm3 · µm/m2 /24h/bar
for the core layer of the multilayer film at 53% RH and 561
cm3 · µm/m2 /24h/bar at 75% RH. These values are widely
higher than these of TPS-AA-Fe film alone. This indicates
that the properties of the TPS-based layer are modified by
the coextrusion with PCL, which suggests interactions between both layers. That can be due to the compatibilisation
effect between PCL and TPS thanks to ascorbic acid. The
coextrusion of the TPS active film with PCL improves its
mechanical properties. Its barrier properties to oxygen remain rather good but slightly increase with water absorp-
Unauthenticated
Download Date | 6/15/17 10:03 PM
Ascorbic acid and iron in mechanical and oxygen absorption properties
tion due to the lack of barrier properties to water of the
multilayer film.
The oxygen absorption capacity of the active films is
characterized by measuring the oxygen rate in bottles containing samples of different films and a saturated salt solution to achieve a humidity rate of 94%. The evolution of the
average oxygen rates as a function of time is presented in
Fig. 5. As obtained previously [9], with the active TPS-AAFe film the oxygen rate decreases to less than 1% in approximately 15 days in the measuring chamber. The oxidation
of ascorbic acid in presence of iron and water consumes all
the oxygen present in the environment of the sample. This
oxygen absorption capacity is very interesting for active
food packaging uses. By elimination of the oxygen around
the food, the packaging will contribute to extend the food
shelf-life. In this case the water absorption is necessary to
the active properties of the film because the absorbed water activates the oxidation of ascorbic acid and the absorption of oxygen by the film.
The coextrusion of TPS-AA-Fe with PCL layers allows
to improve the mechanical properties of this film but it
also results in a slowing down of the oxygen rate decrease
thanks to this film. With the PCL/TPS-AA-Fe/PCL multilayer film at 94% RH, it takes approximately 35 days to
reach an oxygen rate in the measuring chamber under
1%. This slowing down of the oxidation kinetic can be explained by the slowing down of the humidity absorption
due to the PCL layers (as explained at the end of 3.3 section). Some multilayer films based on TPS-AA-Fe have been
perforated on the upper layer in order to facilitate the water and oxygen going through. This led to a little faster oxygen absorption: the oxygen rate in the measuring chamber reaches less than 1% in approximately 30 days. These
results are close to these obtained with a film containing
TPS-AA-Fe matrix blended with 20% (w/w) of PCL [9].
From these measurements we should also notice that
after approximately 20 days delamination of PCL layers begins to occur. The TPS based layer swells as water absorption increases. So the dimensions of the TPS layer change
whereas these of PCL layers do not evolve, leading to a separation of the layers. The active multilayer film presents
interesting oxygen absorption capacity but the kinetic of
absorption must be taken into account for the choice of
the type of food that could be advantageously packaged
in this material. To avoid the delamination the conditions
of use of this material should be limited. Maybe the relative humidity in the package headspace could be limited
at a value lower than 94% RH. The film TPS-AA-Fe without
coextrusion allows to reach less than 1% of O2 in 15 days at
80% RH [9]. But in this case an alternative method should
| 9
Figure 5: Evolution of oxygen rate in the measuring chambers at
94% RH for different films.
be looked for in order to accelerate the reaction of oxygen
absorption.
4 Conclusion
This work demonstrated that ascorbic acid associated with
iron acts as a compatibiliser between starch-based and
PCL layers in coextruded PCL/TPS multilayer films. This
provides a sufficient cohesion between layers and very
good mechanical properties for the multilayer films at ambient humidity (53% RH). The mechanical behavior of the
multilayer film containing the active components is largely
improved in comparison with that of TPS-AA-Fe film alone
and with pure TPS, whereas PCL layers represent only 30%
(w/w) of the multilayer film mass. However, the PCL layers do not efficiently protect the internal TPS-based layer
against humidity. High water absorption of the TPS layer
can lead to a separation of the layers. But the water absorption is necessary to activate the oxygen absorption
and can allow to regulate the humidity rate in the package headspace when the packaged food emits water vapor.
For example this film could be envisaged to package fresh
fruits and vegetables. Those products release water vapor
and are sensible to oxidation, so water and oxygen have to
be removed to increase their shelf-life.
The oxygen scavenging action of the active layer is
slowed down in the multilayer films, with 30 days to absorb all the oxygen present in the measuring chamber
against 15 days without the PCL layers. So coextrusion
of TPS-based active layer with PCL improves mechanical
properties of the film but decreases oxygen absorption kinetic. Water absorption activates and accelerates oxygen
Unauthenticated
Download Date | 6/15/17 10:03 PM
10 | A. Mahieu et al.
absorption but can degrade the material cohesion. The
starch-based film has good barrier properties to oxygen
when it is low hydrated, but when water is absorbed to activate oxygen absorption, O2 barrier properties decrease.
A compromise has to be found between all these properties, depending on the requirements of the food to be packaged. Other alternatives should be investigated to protect
the film from oxygen and water coming from the outside.
For example the upper and lower layers could be differentiated to introduce an oxygen barrier component in the
lower layer and a PCL layer only as the upper layer. However the compatibilising effect of ascorbic acid and iron between plasticized starch and PCL should be remembered
and could be used in other compostable packaging materials without objectives of food preservation.
Acknowledgement: The authors thank the Region Haute
Normandie (France) for financial support.
References
[1]
Liu H., Xie F., Yu L., Chen L., Li L., Thermal processing of starchbased polymers, Prog. Polym. Sci., 2009, 34, 1348-1368.
[2] Averous L., Biodegradable Multiphase Systems Based on Plasticized Starch: A Review, J. Macromol. Sci., Polym. Rev., 2004,
C44, 231-274.
[3] Chaléat C., Halley P.J., Truss R.W., Mechanical properties of
starch-based plastics, in Halley P., Averous L. (Eds.), Starch
Polymers, Elsevier, 2014
[4] Mensitieri G., Di Maio E., Buonocore G.G., Nedi I., Oliviero M.,
Sansone L., et al., Processing and shelf life issues of selected
food packaging materials and structures from renewable resources, Trends Food Sci. Technol., 2011, 22, 72-80.
[5] Sánchez-González L., Cháfer M., Hernández M., Chiralt A.,
González-Martínez C., Antimicrobial activity of polysaccharide
films containing essential oils, Food Control., 2011, 22, 13021310.
[6] Robertson G.L., Active and intelligent packaging, in Robertson
G.L. (Eds.), Food packaging: principles and practice, CRC Press,
2006
[7] Siracusa V., Rocculi P., Romani S., Dalla Rosa M., Biodegradable polymers for food packaging: a review, Food Sci. Technol.,
2008, 19, 634-643.
[8] Fabra M.J., López-Rubio A., Lagaron J.M., Biopolymers for food
packaging applications, in Aguilar De Armas M.R., Román J.S.
(Eds.), Smart Polymers and their Applications, Elsevier, 2014
[9] Mahieu A., Terrié C., Youssef B., Thermoplastic starch films and
thermoplastic starch/polycaprolactone blends with oxygenscavenging properties: Influence of water content, Ind. Crops
Prod., 2015, 72, 192-199.
[10] Follain N., Joly C., Dole P., Bliard C., Mechanical properties of
starch-based materials. I. Short review and complementary experimental analysis, J. Appl. Polym. Sci., 2005, 97, 1783-1794.
[11] Myllymäki O., Myllärinen P., Forssell P., Suortti T., Lähteenkorva
K., Ahvenainen R., et al., Mechanical and Permeability Properties of Biodegradable Extruded Starch/ polycaprolactone Films,
Packag. Technol. Sci., 1998, 1, 265-274.
[12] Li G., Favis B.D., Morphology Development and Interfacial Interactions in Polycaprolactone/ Thermoplastic-Starch Blends,
Macromol. Chem. Phys., 2010, 211, 321-333.
[13] Martin O., Avérous L., Poly(lactic acid) : plasticization and properties of biodegradable multiphase systems, Polymer., 2001,
42, 6209-6219.
[14] Martinez-Abad A., Sanchez G., Fuster V., Lagaron J.M., Ocio
M.J., Antibacterial performance of solvent cast polycaprolactone (PCL) films containing essential oils, Food Control, 2013,
34, 214-220.
[15] Mahieu A., Terrié C., Agoulon A., Leblanc N., Youssef B., Thermoplastic starch and poly(ϵ-caprolactone) blends: morphology
and mechanical properties as a function of relative humidity, J.
Polym. Res., 2013, 20, 1-13.
[16] Ortega-Toro R., Contreras J., Talens P., Chiralt A., Physical and
structural properties and thermal behaviour of starch-poly(ϵcaprolactone) blend films for food packaging, Food Packag.
Shelf Life., 2015, 5, 10-20.
[17] Ortega-Toro R., Collazo-Bigliardi S., Talens P., Chiralt A., Influence of citric acid on the properties and stability of starchpolycaprolactone based films, J. Appl. Polym. Sci., 2015, (in
press) DOI: 10.1002/app.42220.
[18] Martin O., Schwach E., Avérous L., Couturier Y., Properties
of biodegradable multilayer films based on plasticized wheat
starch, Starch/Stärke., 2001, 53, 372-380.
[19] Alix S., Mahieu A., Terrie C., Soulestin J., Gerault E., Feuilloley M.
G. J., et al., Active pseudo-multilayered films from polycaprolactone and starch based matrix for food-packaging applications,
Eur. Polym. J., 2013, 49, 1234-1242.
[20] Yu L., Dean K., Bi L., Polymer blends and composites from renewable resources, Prog. Polym. Sci., 2006, 31, 576-602.
[21] Yu L., Dean K., Yuan Q., Chen L., Zhang X., Effect of compatibilizer distribution on the blends of starch/biodegradable
polyesters, J. Appl. Polym. Sci., 2007, 103, 812-818.
[22] Wu C.-S., Physical properties and biodegradability of maleatedpolycaprolactone/starch composite, Polym. Degrad. Stab.,
2003, 80, 127-134.
[23] Avella M., Errico M. E., Laurienzo P., Martuscelli E., Raimo M.,
Rimedio R., Preparation and characterization of compatibilized
polycaprolactone/starch composites, Polymer., 2000, 41, 38753881.
[24] Lourdin D., Bizot H., Colonna P., “Antiplasticization” in starchglycerol films?, J. Appl. Polym. Sci., 1997, 63, 1047-1053.
[25] Godbillot L., Dole P., Joly C., Rogé B., Mathlouthi M., Analysis of
water binding in starch plasticized films, Food Chem., 2006, 96,
380-386.
[26] Mikus P.-Y., Alix S., Soulestin J., Lacrampe M.F., Krawczak P., Coqueret X., et al., Deformation mechanisms of plasticized starch
materials, Carbohydr. Polym., 2014, 114, 450-457.
[27] Xie F., Liu P., Yu L., Processing of plasticized starch-based materials: state of the art and perspectives, in Halley P., Averous L.
(Eds.), Starch Polymers, Elsevier, 2014
[28] Capron I., Robert P., Colonna P., Brogly M., Planchot V., Starch
in rubbery and glassy states by FTIR spectroscopy, Carbohydr.
Polym., 2007, 68, 249-259.
Unauthenticated
Download Date | 6/15/17 10:03 PM
Ascorbic acid and iron in mechanical and oxygen absorption properties
[29] Greenspan L. J., Humidity fixed points of binary saturated aqueous solutions, Res. Natl. Bur. Stand., Sect. A., 1977, 81A, 89-96.
[30] Buyn Y., Bae H.J., Whiteside S., Active warm-water fish gelatin
film containing oxygen scavenging system, Food Hydrocolloids,
2012, 27, 250-255.
[31] Leblanc N., Saiah R., Beucher E., Gattin R., Castandet M.,
Saiter J.M., Structural investigation and thermal stability of new
extruded wheat flour based polymeric materials, Carbohydr.
Polym., 2008, 73, 548-557.
[32] Couturaud B., Baldo A., Mas A., Robin J. J. J., Improvement of the
interfacial compatibility between cellulose and poly(L-lactide)
films by plasma-induced grafting of L-lactide: The evaluation of
the adhesive properties using a peel test, Colloid Interface Sci.,
2015, 448, 427-436.
| 11
[33] Ortega-Toro R.,Morey I., Talens P., Chiralt A., Active bilayer films
of thermoplastic starch and polycaprolactone obtained by compression molding, Carbohydr. Polym., 2015, 127, 282-290.
[34] Buettner G.R., In the absence of catalytic metals ascorbate does
not autoxidize at pH 7: ascorbate as a test for catalytic metals,
J. Biochem. Biophys. Methods., 1988, 16, 27-40.
[35] Dole P., Joly C., Espuche E., Alric I., Gontard N., Gas transport
properties of starch based films, Carbohydr. Polym., 2004, 58,
335-343.
Unauthenticated
Download Date | 6/15/17 10:03 PM