LWT - Food Science and Technology 54 (2013) 447e455

Contents lists available at ScienceDirect

LWT - Food Science and Technology

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

Physicochemical and mechanical properties of extruded laminates

from native and oxidized banana starch during storage
Yunia Verónica García-Tejeda a, Carlos López-González a, Juan Pablo Pérez-Orozco b,
Rodolfo Rendón-Villalobos a, Alfredo Jiménez-Pérez a, Emmanuel Flores-Huicochea a,
Javier Solorza-Feria a, *, C. Andrea Bastida a
a
Instituto Politécnico Nacional, CEPROBI, Km. 6, Carretera Yautepec-Jojutla, Calle Ceprobi 8, Col. San Isidro, Z.C. 62731 Yautepec, Morelos, Mexico
b
Instituto Tecnológico de Zacatepec, Departamento de Ingeniería Química y Bioquímica, Calzada Tecnológico #27, Apartado postal 45, C.P. 62780 Zacatepec,
Morelos, Mexico

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

Article history: Changes in some mechanical and physicochemical properties of extruded Laminates, made from native
Received 28 November 2012 (NBS) and oxidized banana starch (OBS) with 3 g/100 g of sodium hypochlorite, as affected by storage
Received in revised form time (0e45 days) were evaluated. Micrographs showed cracks and pores with some continuity on the
27 May 2013
surface of both OBS and native banana starch (NBS) laminates. Extruded Laminates of OBS were more
Accepted 28 May 2013
transparent, soluble and homogeneous than those from NBS. Little differences were observed in water
vapor permeability between NBS and OBS Laminates. Laminates solubility decreased with storage time.
Keywords:
The X-ray diffraction from NBS and OBS Laminates, showed similar type B patterns and percent of
Oxidized starch
Extruded laminates
crystallinity. Throughout storage time, an increase in temperature and enthalpy of melting was observed
Physicochemical properties for all Laminates, however, the enthalpy values for OBS Laminates were lower than those of NBS. The
Modified starch tensile strength, percent of elongation and elasticity modulus values of OBS Laminates, were higher than
Musa paradisiaca those of NBS. Overall, OBS might be a suitable raw material to produce extruded Laminates with
adequate functional properties.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction
Nowadays, the indiscriminate use of plastic bags has become
such an increasing environmental pollution problem, that local
governments have passed laws in order to limit their use. Over the
last years, the interest in biodegradable films made from renewable
and natural polymers has increased. Water-soluble polysaccharides
such as starch, cellulose derivatives, alginate and pectin can form
biodegradable and edible films (Garcia, Ponotti & Zaritzky, 2006).
Thus, for example, the substitution of plastic films by short term
degradation ones, mainly those from natural sources, has
been strongly encouraged. The production of films or laminates
(thickness > 1 mm), based on biological materials uses film-
forming agents (e.g., macromolecules as polysaccharides, starch
* Corresponding author. Tel.: þ52 55 57 29 60 00x82543; fax: þ52 55 57 29 60
00x82521.
E-mail addresses: jsolorza@ipn.mx, jsolorzaferia@hotmail.com, j.solorzaferia@
gmail.com (J. Solorza-Feria).
and proteins), a solvent (usually water) and a plasticizer (e.g., pol-
yols as glycerol, sorbitol, ethylene glycol).
Starch consists of two chemically distinct polysaccharides,
namely; amylose and amylopectin, conforming characteristic
structures called granules with both amorphous and crystalline
structures. The amylose is an essentially linear macromolecule of
glucose units linked mainly by a-D-(1-4) bond, although there may
be some a-D-(1-6) bonds. It is distributed mostly in the starch
granule amorphous domains (lamella), with small amounts in the
semicrystalline granule ring. The amylopectin is a branched glucose
polymer, linked mainly by a-D-(1-4) bonds (about 96 g/100 g) and
the remaining 4e6 g/100 g by a-D-(1-6) bonds, conforming the
crystalline lamella (Hermansson & Svegmark, 1996). The propor-
tion of these two polymers and their physical organization inside
the granules, confer singular physicochemical characteristics and
functional properties to starch (Thomas & Atwell, 1999). Starch
granules are semi-crystalline polymeric systems, where the degree
of crystallinity (15e45 g/100 g), is due to short-chain fraction of the
amylopectin arranged as double helices and packed in small crys-
tallites organized in a tri-dimensional structure (Du, MacNaughtan
& Mitchell, 2011).

0023-6438/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.lwt.2013.05.041
448 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455
Banana is a general term involving a number of species or
hybrids in the genus Musa of the family Musacea. Plantains are
generally the larger, more angular starchy fruits of hybrid culti-
vars in the banana family, edible raw when fully ripe, but also
suitable for cooking. They are mainly produced in developing
countries, being of major importance to people in the growing
areas, where a lot of the produce is lost because of poor post-
harvest handling methods. Nonetheless, banana starch has the
potential to be a commodity starch because of its specific prop-
erties and its potential production from low-cost, cull bananas
(Zhang, Whistler, BeMiller &Hamaker, 2005). Various authors
have attempted to produce films from starch isolated from
various sources including maize, potato, cassava, tapioca, flax-
seed and recently banana (Zhang et al., 2005). The most used
techniques for film processing include casting and extrusion,
being the latter a convenient technique for practical purposes.
Starch film extrusion is more complex than that of conventional
polymers due to multi-phase transitions involved during pro-
cessing. There are many works done on films produced by cast-
ing, but further investigation is needed on the physical properties
of those obtained by extrusion.
Starch granules during extrusion are subjected to high shear
stress and temperature, these factors produce the destruction of
granules, elimination of the crystalline structure, at times the
molecular weight reduction, and also, the formation of an
amorphous molten mass (thermoplastic starch polymer, TSP),
which is then pressurized and shaped by the die head (Su et al.,
2009).
Modified starch has been used to replace native starch in
various applications, because of its appropriate functional and
physicochemical properties as a result of its granule modifica-
tion (Kuakpetoon & Wang, 2008). A commonly used method
of starch modification is its oxidation with chemical agents
differing in their oxidation power. Oxidized starch is widely
used in many industries, particularly in applications where film
formation and adhesion properties are desired. The major
application of oxidized starches is as surface sizing agents
and coating binders in the paper industry (Sangseethong,
Termvejsayanon & Sriroth, 2010). Oxidized starch is commonly
produced by reaction of starch with an oxidizing agent under
controlled temperature and pH. Previous studies on corn
starches have found that, when sodium hypochlorite (NaOCL)
has been used, NaOCL concentrations of 2 g/100 g solution or
higher, are enough to produce real changes on such starches
physicochemical characteristics (Kuakpetoon Wang, 2008).
During the oxidation process, hydroxyl groups on starch mole-
cules are oxidized to carbonyl and carboxyl groups, contributing
to improved stability of starch paste. The reaction also causes
degradation of starch molecules resulting in a modified starch
with low viscosity. This allows the use of oxidized starch
in applications where high solid concentration is needed
(Wurzburg, 1986) The addition of plasticizers is necessary to
obtain a TSP by reduction of intermolecular forces, and increase
in the mobility of polar polymer chains. In addition, plasticizers
help to overcome the brittleness of starch films and improve
their flexibility and extensibility. Glycerol has been one of the
most used plasticizers, since it improves film extensibility
(Gurgel, Vieira, da Silva, dos Santos & Beppu, 2011). To define
possible uses and shelf life of these types of materials, studies
on the effect of storage on the physicochemical properties are
needed, but those studies are not available in the literature.
The purpose of this work was to evaluate the changes in some
physical and physicochemical properties of extruded laminates,
made from native and oxidized banana starches, as affected by
storage time.
2. Material and methods
2.1. Materials
Unripe banana (Musa paradisiaca L.) selected with no yellow
color on its peel (about 18 weeks from blossom), was obtained from
the province of Veracruz, Mexico.
2.2. Starch isolation
Native banana starch was obtained as described by González
(2003). This consisted in peeling, cutting and immersing the ba-
nana slices obtained in 0.25 g/100 g citric acid solution. The slices
were blended three times with distilled water (water:banana ratio
1:1) using a Waring blender. The blended fruit pulp was sieved
using an electric sieving machine (Testing Equipment, Model RNU).
Sieves with mesh sizes 40, 100, 200 and 325 (0.425, 0.15, 0.075 and
0.045 mm, respectively), were used one at a time. Each paste
formed in every sieve was washed with water at least three times
(10 parts of water: 1 part of banana pulp) until the wash water was
clean. The pulp cake was eliminated. The filtrate contained starch
and smaller particles suspended. The starch was allowed to pre-
cipitate during 24 h at room temperature (about 25  C). The excess
of water was decanted until getting a solid concentration within the
range of 30e35 g/100 g. The slurry obtained was dried with a Niro
Atomizer spray-dryer (Model 230 EA 11 No. 21, Denmark). The
processing parameters were: inlet temperature of about 140  C;
outlet temperature of about 68  C, with a feeding flow of 14 L/h. The
dehydrated native starch was standardized with a 100 mesh sieve
(0.15 mm) and stored at room temperature in a glass container, till
use for further experiments.
2.3. Starch chemical analysis
For the sake of comparison, the total starch and also the
apparent amylose content (Gilbert & Spragg, 1964), was obtained
with potato starch as standard. For this last component, about
1.5 mg of starch was weighed into an aluminium pan and trans-
ferred to a 50 ml volumetric flask. Then, about 0.5 ml of a solution
NaOH 1 mmol equi/L was added and the mixture simmered during
3 min in a boiling water bath. It was then cooled down and
neutralized with 0.5 ml of 1 mmol equi/L HCl, and 0.07 g of po-
tassium bitartrate was added and diluted with distilled water until
obtaining an approximate volume of 45 ml. Once the bitartrate was
dissolved, 0.5 ml of a solution of iodine (2 mg/ml of Iodine and
20 mg/ml of KI) was added and the volume completed to 50 ml
with distilled water. The solution was mixed and allowed to rest
during 20 min at room temperature, and the absorbance was
measured at 680 nm in a Genesys 5 spectrophotometer (Spectronic
Instruments, USA). The amylose content in the starches was
quantified by interpolation of the absorbance values in a standard
curve. The amylopectin content was obtained by difference.
For total starch content (Goñi, García & Saura-Calixto, 1997), a
50 mg sample was dispersed in 2 mol/equi/L KOH (30 min) and
incubated in a controlled-temperature water bath (60  C, 45 min,
pH 4.75) with amyloglucosidase (Roché No. 102 857, Roche Di-
agnostics, Indianapolis, IN, USA); the glucose hydrolyzate obtained
was measured with a GOD/POD reagent (SERA-PAKÒ Plus, Bayer de
México, SA de CV). Total starch content was calculated as glucose
(mg)  0.9; potato starch was used as standard.
2.4. Starch oxidation
To obtain oxidized banana starch (OBS), native banana starch
(NBS) and sodium hypochlorite solution (NaOCl) with 3 g/100 g
 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455 449

active chlorine was used; the modified method described by Wang taken for each sample, doing each test in triplicate. The measure-
and Wang (2003) was applied. The procedure was as follows: 35e ments were reported using the CIELAB (L*, a*, b*) system. Chro-
45 g/100 g starch slurry was prepared with deionized water. This maticity (C) and hue angle ( h) were evaluated using the following
was set at 35  C by using a magnetic stirrer with a heating system; equations:
the pH was adjusted to 9.5 with a 1 mol equi/L NaOH solution. Then,
50 mL of sodium hypochlorite solution were added at a constant
1=2
rate, slowly and during exactly 30 min, with constant stirring of the C ¼ a*2 þ b*2 (1)
reaction mixture into the flask, using a 1 mol equi/L H2SO4 solution
to maintain the pH at 9.5. The mixture was let to react 50 more min, !
keeping the pH at 9.5 with a 1 mol equi/L NaOH solution. At the 1 b*
end, the pH was set to 7.0 with 1 mol equi/L H2SO4 solution and the
0
h ¼ tan for a* > 0 and b*  (2)
a*
starch separated by decantation. Finally, the remaining oxidized
starch was washed four times with deionized water and dried in a !
convection oven at 45  C for 48 h. 1 b*
0
h ¼ 180 þ tan for a* < 0 (3)
a*
2.5. Laminate production
2.7. Scanning electron microscopy (SEM)
A single screw laboratory extruder (19 mm diameter; ratio
length/diameter of 24) with three heating zones (Beutelspacher, A Scanning Electron Microscope (JEOL JSMP 100, Japan) was
Model 19-24, Mexico, D. F.) was used to produce the laminates used to take SEM micrographs from both the surface (magnification
(thickness > 1.0 mm) from native and oxidized banana starch. The of 500) and from the lateral (magnification of 40) sides of NBS
process consisted in two stages: 1) an optimum mixture obtained by and OBS laminates, which were those more clearly distinguished.
trial and error, of 65 g/100 g starch (either NBS or OBS), 17 g/100 g The laminates were fixed to stainless steel stubs, dehydrated with
glycerol (Merck, Mexico), and 18 g/100 g deionized water, were osmium tetraoxide (OsO4) and covered with a layer of colloidal gold
homogenized thoroughly during about 2 h and let in rest 24 h at of about 20 nm of thickness in the ionizer of metals JEOL. Samples
room temperature (25  C) to stabilize. This mixture was processed at were observed with 5 kV of voltage.
a screw speed of 80 rpm. The selected temperature profile for the
extrusion process was determined from trial and error experiments
2.8. Water vapor permeability
as shown in Table 1. For inlet (zone I), processing (zone II) and outlet
(zone III) areas, respectively. A 1.0 mm diameter nuzzle was used to
The water vapor permeability (WVP) was evaluated using the
obtain the extruded laminates, were cut manually with scissors at
modified method E 96-80, also known as “the test cell” (ASTM,
room temperature, to obtain small fragments as pellets of about
1989). The laminates were previously equilibrated for at least five
5 mm long. The pellets were extruded over again at the same tem-
days in a desiccator at 57 g/100 g RH using a saturated NaBr solu-
perature conditions, at the same screw speed of 80 rpm, but using a
tion. A circular opening with an area of 0.005439 m2 on the test or
nuzzle with a rectangular inside channel of 200 mm width per
permeation cell was covered thoroughly with a laminate of the
0.5 mm thick. All extruded laminates obtained were conditioned at
same shape and area. Dried silica was placed into the cell, making it
25  C and a relative humidity (RH) of about 57 g/100 g, in a desiccator
almost exempt of moisture (about 0 g/100 g RH). An atmosphere of
with a saturated solution of NaBr, before testing. The color and
75 g/100 g RH was obtained with a saturated solution of NaCl out of
morphology were determined in samples kept for at least five days
the permeation cell. Every hour, the gain in weight was monitored
at the above named conditions, while all other physical and physi-
until no further variations were observed. Once the permeation
cochemical tests were undertaken through the storage time (0e45
tests were done, the laminate thicknesses were measured at ten
days). All tests were done at least in triplicate.
different points using a digital micrometer (Mitutoyo, Tokyo, Japan)
and the WVP in g Pa1 s1 m1 was calculated using the following
2.6. Color evaluation equation.

A completely random model was used to choose the surface  

WVTR
points on the laminates to evaluate the color. A universal Milton WVP ¼ d (4)
SðR1  R2 Þ
Roy colorimeter, Color-Mate (USA) with a D65 illuminant and
observation angle of 10 was used to do the test. Five readings were
Where: WVTR is the water vapor transmission rate, calculated
Table 1
from the slope of the straight line (g/s) divided by the cell area (m2).
Temperature program assayed for the laminates extrusion from banana starch. S is the saturated water vapor pressure at the test temperature
(25  C), R1 is the relative humidity (RH) of the desiccator, R2 is the
TIa TIIb( C) TIIIc( C) Comments
( C)
relative humidity of the permeation cell, and d the laminate
thickness (m).
80 100 80 Fail on laminate formation: the material did not flow
uniformly.
80 100 160 Laminate formed with bubbles, hard texture and brown 2.9. Laminates solubility
color.
80 100 100 Laminate formed with bubbles
Laminates of 20  30 mm were stored one week into a dry
80 100 90 Laminate with no bubbles, but rough texture with
wrinkles desiccator with silica gel, until the g/100 g RH of equilibrium was
100 120 90 Smooth and homogeneous laminates almost exempted of moisture. The laminate samples were weighed
a
Temperature in zone I.
(initial dry weight) and immersed in 250 mL beakers containing
b
Temperature in zone II. 80 mL of deionized water. Laminates were kept under constant
c
Temperature in zone III. stirring for 1 h, at either 25 or 80  C. This last temperature was tried,
450 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455
since previous studies have shown that, to dissolve any remains of
oxidized amylose and its complex species, temperatures higher
than 65  C are needed (Lawal, 2004). After soaking, the samples
were dried in an oven at 60  C to constant weight (final weight).
The tests were run in triplicate. The percent of solubility was
calculated as follows:
 
Initial weight dry basis  final weight dry basis
% Solubility ¼  100
Initial weight dry basis
2.10. X-ray diffraction
Samples of laminates at different storage times were cut into
pieces of 15  20 mm and analyzed with a Siemens diffractometer
(Model D5000, Germany), working at a voltage of 30 kV, 20 mA,
CueK radiation, wave length of 1.54  A. The tests were done with the
following parameters: angle interval 3e40 (2q), chart speed was
10 mm 2q with a running rate of 2q min1 and a measuring time of
5 s. The crystallinity degree was calculated as the ratio of the
crystalline area respect to the total area. The software Win PLOTR
(Full Prof Team); a windows tool for powder diffraction patterns
analysis, was employed to draw the graphics.
2.11. Differential scanning calorimetry
With the aid of a differential scanning calorimeter (DSC) (TA
Instruments, Model 2010; New Castle, DE), the melting tempera-
ture (Tf), and the melting enthalpy (DH, evaluated by integrating
the peak area, corresponding to such transition) of the laminates
were measured. The calibration was done with Indium (point of
fusion of 156.4  C; enthalpy of 6.8 cal/g). The base line was achieved
by running a heating program from room temperature (about
25  C) to 250  C. About 2.5 mg of laminate sample was weighed
with a micro balance (Model AD2Z, Perking-Elmer Corp., St Louis,
MO, USA; 0.01 mg of precision). The above mentioned heating
program was run three times.
2.12. Texture evaluation
A texture meter (Stable Micro Systems, Haslernere, UK and
Texture Technologies Corp., Scarsdale, NY., USA), was employed to
measure the tensile strength or tensile fracture (maximum force/
cross sectional area; TF, MPa), percent of elongation at break
(percent of change of the initial length of the sample; %E), and
elastic modulus (EM, MPa), evaluated as the slope of the forcee
deformation plot of the laminates. A cell load of 25.0 kg was used
for testing. A laminate height of 7 cm and a speed of deformation of
0.9 mm/min were set. All samples were prepared according to the
official method ASTM (1995) D-888-95 and were placed five days
into an atmosphere of 57 g/100 g of RH (desiccator with saturated
NaBr) at 25  C before the texture evaluation. Laminate thicknesses
were measured randomly in at least ten different positions of the
specimens with a digital micrometer (thickness gauge Mitutoyo,
Tokyo, Japan).
2.13. Statistical analysis
A one way analysis of variance (ANOVA) with a significance level
of 5% (*P < 0.05), was applied (Montgomery, 1991) to evaluate the
effect of storage on Laminate characteristics. The SigmaStat ver.
2.01 (Systat Software Inc., San José Calif., USA) statistical program
was used.
3. Results and discussion
3.1. Laminates production condition and appearance
Films produced from starches are usually regarded as hydro-
philic materials because of its glycerol content. The oxidized starch,
(5)
a chemical modification of native starch, produces low viscosity at
high solid concentrations and excellent film-forming properties
(Kuakpetoon & Wang, 2001). Table 1 shows that any combination of
extruder inlet temperatures lower than 100  C and processing
temperatures below 120  C, did not produce proper laminates,
probably because no real thermoplasticization of starch was ach-
ieved. Additionally, an outlet temperature higher than 90  C on
both NBS and OBS, produced laminates with folds and bubbles,
owing to flash expansion of water vapor and glycerol inside the
molten material, although in others instances, the flash expansion
is used to produce extruded foam of polymers (Zhang, Zhu, Li & Lee,
2012). The best combination of the three temperatures zones (I, II
and III) for the proper formation of extruded laminates (TPS for-
mation), were 100, 120 and 90  C, respectively. All laminates ob-
tained were standardized to a thickness of about 1.5 mm by
applying a roll on the samples at the end zone of the extruder
during production. Visual examination showed that laminates ob-
tained from OBS were flexible and easy to handle while those from
NBS were rigid, brittle and a bit sticky, probably because of some
phase separation with diffusion of glycerol to the laminate surface.
Overall, laminates from OBS were more homogeneous and more
transparent than those from NBS, due to oxidation and subsequent
leaching out during native starch modification of pigments, lipids;
which are usually forming complex with amylose inside the starch
granule, and proteins (Wang & Wang, 2003).
The laminate homogeneity of OBS might be related to the lower
viscosity of OBS than NBS (about 1.5 times lower, data not shown),
probably the higher viscosity of NBS did not allow a proper flux of
the molten material through the nuzzle, thus, some wrinkles and
bubbles were formed on the laminate surfaces. This behavior of the
extruded laminate samples during the laminate formation process,
was similar to that observed on previous works using a similar
procedure with potato starch. These researches have also found
differences on extruder processing conditions, in cases where the
viscosity of native starch was three to four times as much as that of
oxidized starch (Zhang, Zhang, Wang, Chen & Wang, 2009). Also, it
has been found that the higher the content of amylose, the more
difficult the starches were to be processed by extrusion, spending
twice as much mechanical work than that for waxy starch (Soest &
Essers, 1997; Thuwall, Boldizar & Rigdahl, 2006), being somehow
consistent with this work, where the amylose concentration pre-
sent in NBS (32 g/100 g) was more than twice as much as that of
OBS (14 g/100 g). This tendency has been previously mentioned by
Wang and Wang (2003), who found that amylose was more prone
to depolymerization than was amylopectin at the same hypoclorite
level, related to the linear chemical structure or the random
arrangement of amylose, that makes it more susceptible to oxida-
tive degradation.
3.2. Color
Respect to the laminate color (Table 2), the luminosity L* was
bigger for OBS, than that for NBS, but it was smaller than the one
 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455 451

Table 2
Color parameters of laminates manufactured with NBS and OBS.

Sample L* a* b* h* C*

NBS 30.24  0.92a 1.59  0.46a 2.73  0.27a 1.04  0.13a 3.16  0.08a
OBS 76.69  0.98b 3.1  0.23b 13.42  0.3b 1.34  0.04b 13.77  0.21b

L* e Luminosity; h* e Hue angle; C* e Chromaticity. Mean values of three measurements  standard error. NBS:native banana starch, OBS: oxidized banana starch.

reported previously (L* ¼ 93 0.03) for oxidized starch (0.5e1.5 g/
100 g) extruded at 95  C (Zamudio-Flores, 2005). It seems that the
effect of Maillard reactions caused the decrease in luminosity in
both starches (native and oxidized) (Bekedam, Schols, van Boekel &
Smith, 2006). The color of the laminates obtained were among the
red-yellow color (between the CIELAB a* and b* parameters)
(Table 2). However, OBS laminates had a yellower color than NBS
laminates, because the a* and h* parameters were higher, amid
bigger saturation (C*). This difference in films color could even be
noted visually, being consistent with previous works done previ-
ously (Alanis-López, Pérez-González, Rendón-Villalobos, Jiménez-
Pérez & Solorza-Feria, 2011).
3.3. Micrographs
Concerning the SEM images as seen in Fig. 1, the laminates made
with oxidized starch, showed similar number of cracks and pores
formation on the surface (magnification 100) to laminates made
of NBS. However, on views of lower magnification (40), sample
appearance in laminates from oxidized starch was “smoother” than
in those of native starch. This was probably caused by the presence
of native starch granules which were not fully plasticized during
the extrusion process (Fishman, Coffin, Onwulata & Willett, 2006),
although some continuity was also observed in all laminates made
with both treated and untreated starch, suggesting that these sys-
tems might have the potential to remain accessible to some
biodegradation factors (e.g. enzymes, chemicals, light).
The white arrow on the micrograph of the lateral view of a
typical oxidized starch laminate, showed cracks that are not un-
common during the SEM study and some of them, might be also
produced by cutting effects while conditioning samples. However,
the mentioned cracks could not be particularly signals of the me-
chanical resistance of the oxidized starch laminates. Nonetheless,
some authors (Mali, Grossmann, García, Martino & Zaritzky, 2006)
have associated the smooth zone formations with best mechanical
properties (e.g. larger shear resistance). Besides, it may well be that
glycerol moieties insertion in the oxidized starch laminates, caused
an increase in compactness degree, but the material homogeneity
during its mixing and passing through the extruder might not be
complete, suggesting that a larger residence time in the extruder, to
reach a homogeneous laminate because or real TPS production,
might still be needed.
3.4. Water vapor permeability (WVP)
It is well known that the water vapor permeability (WVP), is the
result of a driving force of water transfer through a film, given by
water chemical potential difference. Its data are suitable to un-
derstand possible mass-transfer mechanisms and soluteemacro-
molecule interactions in degradable films. The WVP of both the
oxidized and native banana starch laminates (Fig. 2), decreased
substantially at all sampling times during the whole storage period,
without reaching an apparent stable minimum value through this
time, a reduction of more than four times as much was observed

Fig. 1. Scanning electron micrographies of laminates made from native banana starch (NBS) and oxidized banana starch (OBS) made by extrusion. The white circles show the
roughness of the laminates surface. The white arrows point at some fractures and pores. (S) ¼ superficial view (100), (L) ¼ Lateral View (40).
452 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455

Fig. 2. Water vapor permeability (WVP) at 25  C of laminates manufacturedby

extrusion with native (NBS, ) and oxidized (OBS, ) banana starches, as function
of storage time. Laminates were previously equilibrated for five days in a desiccator at Fig. 3. Solubility of laminates manufactured with native (NBS, ) and oxidized (OBS,
57% relative humidity, using saturated NaBr solution. Error bars are standard error of ) banana starch, as function of storage time at 25  C. Error bars are standard error
the media. of the media.

from the beginning till the end of storage. This parameter decreased
faster than the rate of decrease of similar starch laminates reported
formerly, when using lower concentrations of chlorine (0.5e1.5 g/
100 g) as an oxidant agent (Zamudio-Flores, 2005). Possibly factors
as the laminate thickness, the percentage of glycerol added, the
effect of the extrusion on the density of the laminate matrix, and
even the microcracks observed on the laminates, affected the WVP
of the specimens. Some researchers have reported previously
similar WVP values for edible starch based films with hydrophilic
nature, to those found in this work (Bertuzzi, Castro Vidaurre,
Armada & Gottifredi, 2007). However, as shown in Fig. 2, a signif-
icant difference was observed solely at the beginning of the storage
time, being the WVP of the NBS laminates higher than that of OBS
samples, this may well be because the oxidation introduced
carboxyl and carbonyl functional groups in the starch molecule,
with some starch depolymerization. These hydrophilic groups
possibly retained more water molecules lowering as a consequence
the WVP (Lawal, 2004; Wang & Wang, 2003). Overall, NBS lami-
nates showed higher decrease in WVP than OBS laminates
throughout the storage time, suggesting that the oxidation process,
did not affect at the same extent as NBS, the diffusion rate of water
during the water vapor permeation phenomenon.
Runsardthong, 2008; Hashimoto & Grosmann, 2003). Besides, the
pores generated during the extrusion, which might well be filled
with water, could be closed partially due to shrinking of the lami-
nates with storage time, creating a physical barrier for water intake.
For both tests at 25 and 80  C, overall, the OBS laminates gained
more water than those of NBS laminates. During the oxidation
process, undertaken under alkaline conditions, starch gelatinized,
resulting thus in higher granules swelling and solubility, which
may well account for the higher solubility of the oxidized sample
(Lawal, 2004). These results suggest an effect attributed to the
oxidation process, that has been shown to take place mainly on the
starch amorphous region, i.e., on the amylose domain; producing
both carbonyl and carboxyl groups in starch structure, increasing
depolymerization and thus, enhancing also its interaction with
water by hydrogen bonding (Wang & Wang, 2003). Besides, during
the extrusion process there might be a better glycerol (plasticizer)
distribution inside the OBS polymeric chains, favoring a higher
solubility than in NBS. Also, an increase in amylose content

3.5. Solubility

The percent of solubility (measured as the gain of water) at 25  C

(Fig. 3) were lower than those at 80  C (Fig. 4), probably because at
the higher temperature, the higher thermal energy involved made
the laminates more soluble by structural weakening of its compo-
nents, while at 25  C, it was solely the water osmotic pressure that
contributed to laminates solubilization (Lawal, 2004). The solubility
values of the laminates manufactured with native and oxidized
starch decreased with storage time at the two temperatures tested,
as shown at 25  C and 80  C. This suggests that a kind of “aging”
process might have occurred in the laminate matrix or, in other
words, a molecular re-arrangement that reduced the free volume,
giving a closer structure and thus, a lower water penetration has
taken place. This event might well diminish the polar character of
the glycerol-starch composite or TPS. Also, it has been shown that Fig. 4. Solubility of laminates manufactured with native (NBS, ) and oxidized (OBS,
as the temperature and pressure of extrusion is increased, the ) banana starch, as function of storage time at 80  C. Error bars are standard error
solubility decreases (Charutigon, Jitpupakdree, Namsree & of the media.
 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455
decreased the index of solubility, as mentioned above; the native
starch used to obtain laminates had higher amylose content (32 g/
100 g) than its oxidized counterpart (14 g/100 g).
3.6. Laminates crystallinity
The X-ray diffraction of laminates manufactured with native and
oxidized starches showed similar patterns at 15 and 45 days, as
seen in Fig. 5. At the first testing time (fifteen days), both NBS and
OBS laminates were mainly amorphous, but with the storage time,
crystallinity peaks were formed. At the second testing time (forty
five days), the percent of crystallinity for the oxidized starch lam-
inates (19.77 g/100 g) was overall, similar to that of the native
starch laminates (20.07 g/100 g). This might be because the overall
starch crystallinity is attributed mainly to amylopectin, which in
the oxidation process is less depolymerised than amylose (Wang &
Wang, 2003). However, these results need further investigation,
since different trends have been reported previously, for example,
the manufacture of banana starch oxidation by double screw
extrusion, produced sheets with higher crystallinity than those of
sheets from native starch (Alanis-López et al., 2011). However,
Wang and Wang (2003) indicated that there were no significant
changes in starch crystallinity even after oxidation with 2 g/100 g
hypochlorite, due to oxidation taking place mainly in the amor-
phous lamellae, rather than in the crystalline lamellae. Thus,
further research is needed to find out if this effect might pass over
films during production.
When comparing the development of crystallinity versus time
of storage, three peaks of wide shape and low intensity are noto-
rious after 45 days of storage in both native and oxidized starch
laminates: at 2q ¼ 12.7, 17.12 and 22.22 respectively. Besides, a well-
Fig. 5. Diffraction patterns of native (NSF, ) and oxidized (OSF, ) banana starch
laminates manufactured by extrusion. a) at 15 and b) 45 days of storage.
453
defined peak (2q ¼ 19.68) is observed. The signal observed at
2q ¼ 17.2 is attributed to the type B crystallinity due to external and
short chains of amylopectin in glycerol (Soest & Essers, 1997). The
diffraction pattern intensities VH type 2q ¼ 12.7, 19.7, and 22.22
might well be the result of recrystallization during storage, of the
remaining amylose in OBS laminates, and that already present in
NBS laminates (Kuakpetoon & Wang, 2006). Mali, Grossmann,
García, Martino & Zaritzky (2002) also found the B-type pattern
in yam starch films, which remained almost the same even after
ninety days storage, with no significant effect in crystallinity from
glycerol addition, suggesting that changes in this property might be
related to starch source.
3.7. Thermal properties
Table 3 shows the transition temperatures (Tf), and enthalpies
(DH) at its endothermic peaks, related mainly to starch crystallites
melting, for laminate samples evaluated as function of the time of
storage. These peak temperatures ranged from 166 to about 217  C,
as an indication of thermal decomposition of the starch bio-
polymers (amylose and amylopectin). Authors like Mali et al.
(2002) mentioned that amylose recrystallization requires temper-
atures higher than 140  C to be detected by DSC, being consistent
with these results and those of XRD previously shown. The fact that
no signals were detected before the above mentioned temperature
range, suggests that starch gelatinization during film obtainment
was complete. Chang, Cheah & Seow (2000) have suggested that
recrystallization of amylose and perhaps to a lesser extent of
amylopectin, may take place during the film-forming process, an
event that is known to be time-dependent and is likely to occur
during storage. As time passed out, an increase in the temperature
an enthalpy of melting was observed for both NBS and OBS lami-
nates, being the melting temperature of the OBS laminates, higher
than those of the NBS solely within the first fifteen days of films
manufacture, when both NBS and OBS laminates were mainly
amorphous; but changing this trend from thirty days storage on-
wards. Betuzzi et al. (2007), found that the final degree of crystal-
linity in starch films depends on the ability of amylose chains to
form crystals, as well as the mobility of the chains during the
crystallization process. This suggests that the higher amylose con-
tent in NBS might be responsible of the higher melting temperature
as the storage time increased. However, the enthalpy (DH) values
associated with melting of the crystalline phase, were lower for
OBS laminates than those of NBS laminates, possibly because of the
degradation of the crystalline lamellae expected in OBS laminates
(Betuzzi et al., 2007). Thus, it is notorious that the oxidation process
modified the molecular arrangement in a form that it weakened the
starch structure, but rendered it more stable, because despite of the
transition (melting) temperature being higher for NBS laminates
Table 3
Thermal properties of the NBS and OBS laminates during storage*.
Samples Tf ( C) DH (J/g)
NBS0 188.55  0.72a 42.78  0.35a
NBS 15 194.51  0.59b 87.55  1.97b
NBS 30 215.48  0.54c 96.00  1.21c
NBS 45 216.92  0.50c 115.08  1.98d
OSF0 200.50  0.21d 39.79  0.62a
OSF15 203.23  0.46d 62.30  2.5e
OSF30 211.88  0.85e 75.712  0.91f
OSF45 216.66  0.33c 83.86  0.92b
*mean values of three measurements  standard error. 0, 15, 30 and 45 are the
storage days. Same suffixes in the same column indicate no significant differences
(a ¼ 0.05). Tf ¼ transition temperature, DH ¼ transition or melting enthalpy, NBS:
native banana starch; OBS: oxidized banana starch.
454 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455
Fig. 6. Tensile strength of starch laminates made from native (NBS, ) and oxidized
(OBS, ) banana starches. Error bars are standard error of the media.
than those from OBS, the amount of energy (enthalpy) needed for
this transition was lower, a trend that has also been observed in
previous works (Alanis-López et al., 2011; Jouppila & Ross, 1997).
3.8. Laminates texture
Respect to the mechanical properties; tensile strength (TS, MPa)
(Fig. 6), percent of elongation (E, %) (Fig. 7), and elasticity modulus
(EM, slope of forceedeformation plot, MPa) (Fig. 8); of laminates
made with oxidized starch, was bigger in all tests than those of the
native starch: being consistent with the results of Alanis-López
et al. (2011), where it was postulated that the processing temper-
ature essayed, probably affected the oxidized starch laminates
mechanical resistance. The relationship among TS vs. E and EM
varied conversely with time of storage; i.e., as the %E decreased,
both the TS and the EM increased. Laminates from OBS increased its
mechanical properties (TS and EM), respect to those of NBS,
possibly because of the oxidation process itself (e.g., change from
hydroxyl to carbonyl groups, increasing hydrogen bonds), that
Fig. 7. Percent of elongation of starch laminates made from native (NBS, ) and
oxidized (OBS, ) banana starches. Error bars are standard error of the media.
Fig. 8. Modulus of elasticity of starch laminates made from native (NBS, ) and
oxidized (OBS, ) banana starches. Error bars are standard error of the media.
enhanced crosslinks formation in starch network, increasing the
strength of those samples. This behavior has been reported previ-
ously for potato starch laminates and it was attributed also to the
increment of crystallinity (Soest & Essers, 1997), besides, the lower
%E shown by NBS laminates with storage time, is consistent with its
brittle texture mentioned above. Although further research is
needed to understand the changes in laminates from oxidized
starch during longer storage times, this study have shown that,
from its overall functional properties, oxidized banana starch might
have the potential of application in laminates production.
4. Conclusion
Laminates from OBS showed clearer color (higher luminosity)
than those from NBS. Superficial views of laminates (SEM) from
OBS were “smoother” than those of NBS. However, in lateral views,
both laminates made from OBS and NBS showed cracks and pores
formation on the surface. No significant differences were observed
in WVP values between NBS and OBS laminates. The percent of
solubility of the laminates from both NBS and OBS decreased with
storage time, irrespective of the measuring temperature. Similar
type B patterns and percent of crystallinity were observed in lam-
inates from both NBS and OBS. As time passed out, an increase in
the temperature an enthalpy of melting was observed for both NBS
and OBS laminates, but the DH values for OBS laminates were lower
than those of NBS specimens. The strength of laminates made with
OBS, was higher than those of NBS in all tests. The relationship
among TS vs. %E and EM varied conversely with storage time.
Acknowledgments
The authors are grateful to Instituto Politécnico Nacional (Pro-
jects 20121051 and 20131083), CONACYT (Project 60565-Z), SNI,
COFAA, EDI and SIP- in Mexico.
References
Alanis-López, P., Pérez-González, J., Rendón-Villalobos, R., Jiménez-Pérez, A., &
Solorza-Feria, J. (2011). Extrusion and characterization of thermoplastic starch
sheets from “Macho” banana. Journal of Food Science, 76(6), E465eE471.
 Y.V. García-Tejeda et al. / LWT - Food Science and Technology 54 (2013) 447e455
ASTM. (1989). Standard test methods for water vapor transmission of materials in
sheet form. E96e80. Philadelphia: ASTM International.
ASTM. (1995). Standard test methods for tensile properties of thin plastic sheeting,
methods D-6388 M 93 and D-888-95a. Philadelphia: American Society for
Testing and Materials.
Bekedam, E. K., Schols, H. A., van Boekel, M. A. J. S., & Smith, G. J. (2006). High
molecular weight melanoidins from coffee brew. Journal of Agriculture and Food
Chemistry, 54(20), 7658e7666.
Bertuzzi, M. A., Castro Vidaurre, E. F., Armada, M., & Gottifredi, J. G. (2007). Water
vapor permeability of edible starch based films. Journal of Food Engineering, 80,
972e978.
Chang, Y. P., Cheah, P. B., & Seow, C. C. (2000). Plasticizingeantiplasticizing effects of
water on physical properties of tapioca starch films in the glassy state. Journal of
Food Science, 65(3), 445e451.
Charutigon, C., Jitpupakdree, J., Namsree, P., & Runsardthong, V. (2008). Effects of
processing conditions and the use of modified starch and monoglyceride on
some properties of extruded rice vermicelli. LWT-Food Science and Technology,
41(4), 642e651.
Du, X., MacNaughtan, B., & Mitchell, J. R. (2011). Quantification of amorphous
content in starch granules. Food Chemistry, 127(1), 188e191.
Fishman, M. L., Coffin, D. R., Onwulata, C. I., & Willett, J. L. (2006). Two stage
extrusion of plasticized pectin/poly(vinyl alcohol) blends. Carbohydrate Poly-
mers, 65(4), 421e429.
Garcia, M. A., Ponotti, A., & Zaritzky, N. E. (2006). Physicochemical, water vapor
barrier and mechanical properties of corn starch and chitosan composite films.
Starch/Stärke, 58, 453e463.
Gilbert, G. A., & Spragg, S. P. (1964). Iodometric determination of amylose. In
R. I. Whistler (Ed.), Methods in carbohydrate chemistry (pp. 168e169). New York:
Academic Press.
Goñi, I., García, A., & Saura-Calixto, F. (1997). A starch hydrolysis procedure to
estimate glycemic index. Nutrition Research, 17(3), 427e437.
González, A. E. (2003). Obtención de almidón a partir de plátano completo (Musa
paradisiaca L. BSci. Thesis. México: Universidad Tecamac.
Gurgel, M., Vieira, A., da Silva, M. A., dos Santos, L. O., & Beppu, M. M. (2011).
Natural-based plasticizers and biopolymer films: a review. European Polymer
Journal, 47(3), 254e263.
Hashimoto, J. M., & Grosmann, M. V. E. (2003). Effects of extrusion conditions on
quality of cassava bran/cassava starch extrudates. International Journal of Food
Science & Technology, 38(5), 511e517.
Hermansson, A. M., & Svegmark, K. (1996). Developments in the understanding of
starch functionality. Trends in Food Science & Technology, 7, 345e353.
Jouppila, K., & Roos, Y. H. (1997). The physical state of amorphous corn starch and its
impact on crystallization. Carbohydrate Polymers, 32(2), 95e104.
Kuakpetoon, D., & Wang, Y. (2001). Characterization of different starches oxidized
by hypochlorite. Starch/Stärke, 53(5), 211e218.
455
Kuakpetoon, C. A., & Wang, Y. (2006). Structural characteristics and physicochem-
ical properties of oxidized corn starches varying in amylose content. Carbohy-
drate Research, 341(11), 1896e1915.
Kuakpetoon, C. A., & Wang, Y. (2008). Locations of hypochlorite oxidation in corn
starches varying in amylose content. Carbohydrate Research, 343(1), 90e100.
Lawal, S. O. (2004). Composition, physicochemical properties and retrogradation
characteristics of native, oxidized, acetylated and acid-thinned new cocoyam
(Xanthosoma sagittifolium) starch. Food Chemistry, 87, 205e218.
Mali, S., Grossmann, M. V. E., García, M. A., Martino, M. N., & Zaritzky, N. E. (2002).
Microstructural characterization of yam starch films. Carbohydrate Polymers,
50(4), 379e386.
Mali, S., Grossmann, M. V. E., García, M. A., Martino, M. N., & Zaritzky, N. E. (2006).
Effects of controlled storage on thermal, mechanical and barrier properties of
plasticized films from different starch sources. Journal of Food Engineering,
75(4), 453e460.
Montgomery, D. C. (1991). Design and analysis of experiments. New York: John Wiley
and Sons.
Sangseethong, K., Termvejsayanon, N., & Sriroth, K. (2010). Characterization of
physicochemical properties of hypochlorite and peroxide oxidized cassava
starches. Carbohydrate Polymers, 82(4), 446e453.
Soest, J. J. G., & Essers, P. J. J. (1997). Influence of amylose-amylopectin ratio on
properties of extruded starch plastic sheets. Journal of Macromolecular Science,
Part A: Pure and Applied Chemistry, 34(9), 1665e1689.
Su, B., Xie, F., Li, M., Corrigan, A. P., Yu, L., Li, X., et al. (2009). Extrusion processing of
starch film. International Journal of Food Engineering, 5(1), 1e12.
Thomas, D. J., & Atwell, W. A. (1999). Starches (handbook series). St. Paul, MN:
American Association of Cereal Chemists.
Thuwall, M., Boldizar, A., & Rigdahl, M. (2006). Extrusion processing of high amylose
potato starch materials. Carbohydrate Polymers, 65(4), 441e446.
Wang, Y. J., & Wang, L. (2003). Physicochemical properties of common and waxy
corn starches oxidized by different levels of sodium hypochlorite. Carbohydrate
Polymers, 52(3), 207e217.
Wurzburg, O. B. (1986). Modified starches: Properties and uses. Boca Raton, Florida:
CRC Press.
Zamudio-Flores, P. B. (2005). Elaboración de películas degradables de almidón de
plátano, evaluación de sus propiedades mecánicas y de barrera. MSci. Thesis.
Yautepec, Morelos. México: Centro de Desarrollo de Productos Bióticos-IPN.
Zhang, P., Whistler, R. L., BeMiller, J. N., & Hamaker, B. R. (2005). Banana starch:
production, physicochemical properties, and digestibilityda review. Carbohy-
drate Polymers, 59(4), 443e458.
Zhang, Y.-R., Zhang, S.-D., Wang, X.-L., Chen, R.-Y., & Wang, Y. Z. (2009). Effect of
carbonyl content on the properties of thermoplastic oxidized starch. Carbohy-
drate Polymers, 78(1), 157e161.
Zhang, C., Zhu, B., Li, D., & Lee, L. J. (2012). Extruded polystyrene foams with bimodal
cell morphology. Polymer, 53(12), 2435e2442.