Improvement of starch films for food packaging through a three-principle approach: Antioxidants, cross-linking and reinforcement

Carbohydrate Polymers 250 (2020) 116828

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Improvement of starch films for food packaging through a three-principle
approach: Antioxidants, cross-linking and reinforcement

T

Carolin Menzel*
Division of Glycoscience, Department of Chemistry, School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH Royal Institute of Technology, AlbaNova
University Centre, Stockholm, Sweden

ARTICLE INFO

ABSTRACT

Keywords:
Potato starch
Citric acid
Molecular weight
DPPH radical scavenging assay
Compression molding
Molar mass distribution
Crystallinity

This study uses sunflower hulls, a by-product from the sunflower snack industry, to recover both, valuable
phenolic compounds and cellulose fibers, for the production of antioxidant reinforced starch films as potential

food packaging material. The phenolic extract provided antioxidant properties to the films with EC50 values of
89 mg film/mg DPPH. The cellulose fibers reinforced the starch films with a threefold increase in Young´s
modulus. Furthermore, citric acid was added to induce cross-linking of the starch polymers and improve film
integrity. The addition of citric acid induced both, starch polymer hydrolysis and cross-linking, seen in a shift in
chain-length distribution after debranching with iso-amylase. This is the first study that focuses on a threeprinciple approach to improve edible starch films, and follows UN goals on sustainability to reduce waste and
increase value in by-products as a step forward to functionalize packaging material.

1. Introduction
According to the Plastics Europe Market Research Group, the global
plastic production reached about 350 million tonnes in 2017, of which
about 8 million tonnes were calculated to reach the ocean with an
predicted increase by an order of magnitude by 2025 (Jambeck et al.,
2015). Plastics are mainly used for packaging (about 40 %), however,
only 50 % of all plastics produced are considered to be disposable and
only 9 % are recycled (UNenvironmentprogramme, 2019). During the
last two decades, many efforts have been made to find renewable,
biodegradable and non-toxic alternatives to the conventional plastics
like polyethylene (PE), polypropylene (PP), polystyrene (PS) or polyethylene terephthalate (PET). Natural polymers like proteins, polysaccharides or lipids offer a great potential to replace synthetic plastics.
Starch is a biopolymer well known for its biocompatibility, degradability, availability and can easily be converted into a thermoplastic
material. Starch occurs naturally in form of semi-crystalline granules
and consists of the two main polymers amylose and amylopectin, which
contribute to around 25 % and 75 %, respectively. Both macromolecules are built up by chains of α- (1–4) linked D-glucose monomers
with molar masses up to around 106 Da for amylose and 109 Da for
amylopectin. Amylose is considered to be mainly built up by linear
glucose chains with only few side branches, whereas amylopectin exhibits (1–6) α-linked branching points that built up a repetitive cluster

⁎

structure (Pérez & Bertoft, 2010). In order to use starch for packaging
applications, starch has to be plasticized and/or physico-chemical

modified to overcome its hydrophilic character. Chemical modification
can include etherification, esterification or cross-linking reactions,
preferably with food-grade additives in case of food packaging applications. Citric acid has been shown to work as excellent plasticizer and
cross-linker to improve starch films in terms of their water sensitivity,
thermal stability and tensile strength (Menzel et al., 2013; Olsson,
Menzel et al., 2013; Seligra, Medina Jaramillo, Famá, & Goyanes,
2016). Hence, citric acid cross-linked starch films are considered suitable food packaging materials. Another way to improve starch-based
films is the addition of natural fibers as reinforcing component for
thermoplastic materials. Starch-based films with cellulose fibers
showed increased tensile strength and lower water vapor permeability
(Müller, Laurindo, & Yamashita, 2009; Wilpiszewska & Czech, 2014).
Another approach to improve food packaging include the concept of
active packaging, in which an active compound added to the matrix
enhances properties such as antioxidant or antimicrobial activity and
hence, improve the shelf-life of a product (Valdés, Mellinas, Ramos,
Garrigós, & Jiménez, 2014). Especially natural compounds with antibacterial and antioxidant properties gained attention to replace synthetic additives such as butylated hydroxytoluene (BHT). Several studies showed the successful incorporation of antioxidants into starch
films (Luchese, Uranga, Spada, Tessaro, & de la Caba, 2018; Menzel,

Corresponding author at: KTH Royal Institute of Technology, Roslagstullsbacken 21, SE-10044, Stockholm, Sweden.
E-mail address:

/>Received 21 March 2020; Received in revised form 24 June 2020; Accepted 23 July 2020
Available online 01 August 2020
0144-8617/ © 2020 The Author. Published by Elsevier Ltd. This is an open access article under the CC BY license ( />

Carbohydrate Polymers 250 (2020) 116828

C. Menzel

González-Martínez, Chiralt, & Vilaplana, 2019; Menzel, GonzálezMartínez, Vilaplana, Diretto, & Chiralt, 2019)

The recovery of active compounds and biopolymers from food waste
and agro-industrial by-products is a current target of sustainable functional materials and of much interest in recent research (BenitoGonzález, López-Rubio, & Martínez-Sanz, 2019; Chiralt, Menzel,
Hernandez-García, Collazo, & Gonzalez-Martinez, 2020; Galanakis,
2012; Valdés et al., 2014), and in line with the European Commission
Horizon 2020 work program on circular economy and the United Nations Sustainable development goals set on food waste reduction
(Economic & Affairs, 2018).
In this study, three approaches were combined to produce innovative starch films for food packaging with improved properties.
Firstly, citric acid was used as cross-linking and plasticizing agent in
starch films to decrease water sensitivity and improve film integrity.
Secondly, cellulose fibers were added to reinforce the starch network
and increase film strength and decrease water vapor permeability. And
thirdly, an antioxidant extract was added to enhance the antioxidant
properties of the starch films. Furthermore, the importance of this study
lies in the choice of active compounds and fiber origin. Both, the cellulose fibers and antioxidant compounds were extracted from an agroindustrial by-product, sunflower hulls, which is currently regarded as
waste and is underutilized. Processing conditions that simulate industrial parameters were chosen to produce the films, hence, meltblending and compression-molding were studied.

Table 1
Film sample abbreviation including film composition [g] of starch films with
cellulose fibers (Ce), citric acid (Ca) and antioxidant extract (A).
Sample

Starch

Glycerol

Cellulose fibers

Ce
Ca
CeCa

CaA
CeCaA
Ref S
Ref A

40
40
40
40
40
40
40

10
10
10
10
10
10
8

1
1
1

Citric acid

Antioxidant Extract

3

3
3
3

2
2
2

treatment and bleaching were repeated two more times until the material was completely white. The yield of extracted cellulose fibers was
about 27 % by weight. The purity was determined after sulfuric acid
hydrolysis and glucose determination using HPAEC-PAD analysis of
neutral sugars.
2.3. Starch film forming process: melt blending and compression molding

This study is a continuation of previous work with the aim to further
improve the properties of starch films. Antioxidant starch films by
Menzel, González-Martínez, Chiralt et al. (2019) showed low extensibility and remaining high water vapor permeability. Hence, citric
acid cross-linking and cellulose fiber addition targets to improve these
properties. Two reference samples from the previous study by the same
authors were included in to better interpret and compare the results
(Ref A as starch film with antioxidant extract and Ref S with is only
starch-glycerol film).

For each film, 40 g of potato starch was blended with 10 g glycerol
(Table 1). In case of citric acid addition, 3 g crystalline citric acid was
blended with 10 g glycerol and then mixed with the starch powder.
Same procedure was applied to the antioxidant extract, where 2 g of the
extract were added directly to the glycerol and then mixed with the
starch. The cellulose fibers (1 g) were added directly to the starch
powder and then blended with glycerol. The pre-mixed blends were

mixed using a using a Haake PolyLab QC internal mixer (Thermo Fisher
Scientific, Germany) and heated to 160 °C for 7 min at 50 rpm and
torque-time curves were recorded (Supplementary Fig. S1). The blend
was milled using a mill (Moulinex A320R1, 700 W, France) and then
the fine powder was conditioned at 53 % RH at room temperature for 7
days before film molding. The compression-molding was carried out
according to Menzel, González-Martínez, Chiralt et al. (2019). The
starch films were stored at 33 % RH at room temperature before further
analysis.

2.1. Material and chemicals

2.4. Physico-chemical characterization of starch films

Sunflower hulls were kindly provided by Grefusa (Alzira, Spain).
The hulls were washed, dried and milled. A phenolic extract was extracted with aqueous methanol and freeze-dried as described elsewhere
(Menzel, González-Martínez, Chiralt et al., 2019). In brief, milled sunflower hulls were extracted with aqueous methanol at room temperature and the extract was freeze-dried. All chemicals and reagents were
of analytical grade if not further described and purchased from SigmaAldrich (USA). For starch films production, potato starch was used,
which was purchased from Roquette (France) and had an amylose
content of 27 %. Glycerol, sodium carbonate, methanol and ethanol
were from purchased from PanReac Quimica (Spain). Iso-amylase was
purchased from Megazyme (E.C 3.2.1.68 from Pseudomonas sp., 180 U/
mg).

Molar mass and chain length distribution of debranched starch films.
The molar mass distribution of the starch films dissolved in DMSO/LiBr
0.5 % (w/w) was measured according to Vilaplana and Gilbert (2010)
using same size-exclusion parameters.
The chain length distribution of starch was studied after debranching with iso-amylase. About 5 mg starch film was dispersed in 4.5
mL distilled water (1 h in boiling water bath). To the cold dispersion 10

μL 10 ppm sodium azide and 0.5 mL acetate buffer (0.1 M, pH 3.5) was
added. Iso-amylase (25 μL, 20 U/ ml, 4 h at 37 °C) was added and left at
37 °C for 4 h. Afterwards, the solution was precipitated with 25 mL of
absolute ethanol and centrifuged to discard the supernatant. The precipitant was dissolved in DMSO/LiBr 0.5 % (w/w) for 2 h at around 60
°C and injected to a SEC-MALLS/DRI system and molar mass was
measured using a standard calibration with pullulan standards with
molecular weight of 342–708,000 Da. The degree of polymerization
was calculated using the molar mass of anhydroglucose (162 Da).
Starch film appearance and microstructure. Digital pictures of films as
well as field emission scanning electron microscope (FESEM, ZEISS
ULTRA 55 model, Germany) images of cross-sections of each film (1.5
kV acceleration voltage) and cellulose fibers. Therefore, films were
previously dehydrated, cryo-fractured and then gold-coated. Thickness
of films was measured at six points using a digital electronic micrometer with an accuracy of 0.001 mm (Palmer model COMECTA,
Barcelona).
Transparency and color. Color measurement were carried out using a
MINOLTA spectrocolorimeter (Model CM-3600d, Tokyo, Japan) as

2. Material and methods

2.2. Extraction of phenolic extract and cellulose fibers from sunflower hulls
The extraction of the phenolic compounds was in accordance to
Menzel, González-Martínez, Chiralt et al. (2019). Afterwards, the dried
and methanol-extracted sunflower hulls were treated with alkaline 2 %
H2O2 solution at a sample:solvent ratio of 1:10 (w/v) and heated under
stirring to 60 °C for 5 h to extract the hemicellulose and lignin part of
the material, which was used in another study. The peroxide mixture
was filtered under vacuum and washed with distilled water until pH 7
was reached. Afterwards, the material was alkali treated with 4 % (w/
v) NaOH solution at a ratio 1:20 (w/v) under reflux for 3 h, and finally

bleached using 1.7 % (w/v) NaClO2 under reflux for 3 h. The alkali
2


Carbohydrate Polymers 250 (2020) 116828

C. Menzel

described previously (Menzel, González-Martínez, Chiralt et al., 2019).
Three measurements on three films were taken. The CIELab color coordinates (illuminant D65, observer 10°) were obtained from the reflectance of an infinitely thick layer of the material (Hutchings, 1999).
The transparencies of the films at wavelengths ranging from 400 to 800
nm were investigated.
Water vapor permeability. All films were pre-conditioned at 53 % RH
at 23 °C before measurement. Water vapor permeability was determined at 53 %–100 % RH gradient according to the ASTM E96
Standard method (cup method), including the corrections for the air
gap by Gennadios, Weller, and Gooding (1994). The measurement was
carried out in duplicate.
Moisture content and swelling in water. The moisture content of films
conditioned at 33 % RH were determined gravimetrically by drying at
60 °C for 48 h using vacuum until constant weight. The swelling was
measure by immersing a 20 × 20 mm piece of dried film in 15 mL of
water for 24 h at room temperature. Afterwards, the films were wiped
of and weight was measured for the swelling determination.
Mechanical properties. Six to nine replicates of each film formulation
were used to determine tensile properties following the ASTM standard
method (D882.ASTM D882, 2001). Therefore, the conditioned films
were cut into 25 mm x 80 mm pieces, mounted into a Universal testing
machine (Stable Micro System TA, XT plus, Haslemere, England) and a
stretching of 50 mm/ min was applied. The stress-strain curves were
recorded and tensile strength, elongation at break and Young´s modulus

was calculated.
Thermogravimetric analysis. A thermogravimetric analyzer from
Mettler Toleda (TGA/SDTA 851e, Switzerland) was used to heat film
samples from 25 °C to 600 °C at a heating interval of 10 K/ min and to
analyze the onset and peak temperature as well as the loss of weight at
this point. The measurements were done in triplicates.
X-ray diffraction (XRD) analysis. Crystalline structure was characterized by X-ray diffraction (XRD) measurements in air at room
temperature. The patterns were recorded in the range of 5 and 60 2θ
angles with a PANalytical X′Pert PRO diffraction system operated at 40
mA, using CuKα radiation (λ =1.5418 Å). Based on the recorded diffractogram, a simple crystallinity index (Xc) was obtained based on the
method described by Hulleman, Kalisvaart, Janssen, Feil, and
Vliegenthart (1999) using the following Eq. (1):

Xc =

Hc
,
Hc + Ha

3.2. Compression molding of films
All films were easy to process using melt blending and subsequent
compression molding. The torque-time curves during the melt-blending
of the starch mixtures are shown in supplementary Fig. S1. As expected,
there was in increase in torque, which is attributed to increasing internal pressure and viscosity. Starch granules were fractured and
swelled. Then the torque decreases which indicates the agglutination of
starch as studied in detail by Castaño et al. (2017) until a constant
molten steady state was reached. The addition of citric acid resulted in
lower torque, which could be connected to its plasticizing effect but
also partial hydrolytic action during the heat and shear treatment as
discussed below in detail. In contrast, the addition of cellulose fibers

increased the maximum torque. The changes during the melt-blending
have not been studied in detail for the here described starch blends and,
hence, could help in future to optimize the thermo-plasticization process in future applications and industrial up-scaling.
3.3. Appearance of films and optical properties
All starch formulations formed homogenous, transparent and consistent films as shown in the digital images in Fig. 2. The thermocompression conditions were suitable to produce alone-standing films
with a thickness of 0.15−0.19 mm (Table 2), which decreased when
citric acid was added.
The addition of antioxidant extract resulted in slightly yellowbrownish colored films as seen in higher a* and b* color coordinates
(supplementary Table S1), and slightly lower internal transmittance
(lower Ti values, Table 2). The formation of color is due to the light
absorbance of phenolic compounds in the antioxidant extract. The addition of citric acid did not negatively affect the transmittance of the
films, but rather seemed to improve it (Ca film with Ti of 85.8 %)
compared to starch reference films with only glycerol (Ref S of 83.7 %).
Higher values of Ti indicate greater homogeneity. The incorporation of
cellulose fibers had no negative effect on the optical properties.
3.4. Molar mass and chain length distribution using SEC
Results of the molar mass distribution using SEC-MALLS/DRI are
given in Table 3 as weight-average molecular weight Mw, numberaverage molecular weight Mn and polydispersity D. In general, the Mw
decreased drastically in starch films when citric acid was added, which
has been reported previously (Menzel et al., 2013; Shi et al., 2007). The
low Mw in all films indicates extensive hydrolysis of glycosidic bonds in
starch. Furthermore, starch films were debranched using iso-amylase to
determine changes in the chain-length distribution of the starch films
during processing. Fig. 3 showed the typical bimodal distribution of the
reference starch film (Ref S) with amylopectin branches at DP 26 and
48, as well as the amylose branches at DP > 100. Starch-glycerol films,
cellulose fiber and antioxidant extract incorporation slightly changed
the profile towards shorter chains of amylose, but did not affect the
amylopectin bimodal distribution (Menzel, González-Martínez, Chiralt
et al., 2019). However, citric acid incorporation into the starch films

significantly affected the starch structure and showed a monomodal
distribution of chains (Ca, CeCa, CaA, CeCaA in Fig. 3), i.e. the bimodal
branch chain length distribution of amylopectin was not detectable but
shifted towards higher chain length with DP 69–78 and a shoulder at
DP 180. In addition, the amylose peak seemed to vanish or shift completely. Citric acid is known to hydrolyze glycosidic bonds in starch
films when prepared by melt blending (Shi et al., 2007), but at the same
time form ester bonds with starch. If starch had only been hydrolysed,
part of the branch chain-length population would have moved towards
lower values, however, longer chains with a maximum DP at around 75
indicate that intermolecular cross-linkages of starch by citric acid were
formed. Iso-amylase cleaves specifically at the α- (1-6) glycosidic linkages in starch. Hence, chains with a DP 70 might be a result of two

(1)

where Hc and Ha are the intensities for the crystalline and amorphous
profiles with typical baselines at a value of 2Ɵ between 17° and 18°
(Supplementary Fig. S2).
In-vitro antioxidant activity of starch films. A DPPH radical assay was
used to determine the antioxidant activity of the films according to
Menzel, González-Martínez, Chiralt et al. (2019). In brief, films were
dispersed in water for at least 12 h and a filtered aliquot of the solution
was used for the DPPH* assay. Absorbance was read at 515 nm using a
spectrophotometer (Evolution 201 VisibleUV, ThermoScientific, Germany). The measurements were carried out in triplicates. The results
were expressed as efficient concentration of antioxidant in the raw
material to decrease the initial DPPH* concentration by 50 % (EC50, mg
film/ mg DPPH*).
3. Results and discussion
3.1. Extraction of cellulose fibers and characterization
One target of the study was to recover cellulose as biopolymer from
the agro-industrial by-product sunflower hulls. The extraction of cellulose fibers using alkaline extraction and bleaching was successful and

resulted in a yield of 27 % and a purity of 89 %. Cellulose fibers were
white and characterized by FESEM and particle size distribution as
shown in Fig. 1.
3


Carbohydrate Polymers 250 (2020) 116828

C. Menzel

Fig. 1. FESEM image of extracted cellulose fibers after alkaline extraction and bleaching. Digital image of cellulose material in upper right corner and particle size
distribution of cellulose fibers in lower right corner.

adjacent branches being esterified by the tricarboxylic acid. The
shoulder peak with DP 180 was expected to originate from long chain
amylose hydrolyzed by the competing acid action of citric acid during
the thermal processing of melt blending and compression molding as
has been reported before (Menzel et al., 2013; Shi et al., 2007). In
addition, starch films with only added citric acid showed lowest branch
chain lengths compared to starch-citric acid films with cellulose fibers
and antioxidant extract due to the dilution factor and probably steric
hindrance.

showing very low crystallinity due to the complete disruption and gelatinization of starch granules during film processing and the storage
conditions at 33 % RH (van Soest, Hulleman, de Wit, & Vliegenthart,
1996).
Only slight signals of typical B-type and V-type crystalline structure
were observed. The small amounts of detected Vh-type structures (Ref S
and Ref A) are linked to amylose crystallization into single helical
structure. Interestingly, starch-glycerol films with antioxidant showed

predominant Vh-type crystallinity, which suggested that the phenolic
compounds of the antioxidant extract might favored crystallization of
amylose into single helical structure. The formation of B-type crystallinity originates from the recrystallisation of amylose and the amyloseinduced recrystallisation of amylopectin by co-crystallizing into the
same B-type lattice (Hulleman et al., 1999). The B-type crystallinity
indexes Xc for values of 2Ɵ between 17° and 18° are between 0.46 and
0.03 (Table 3). The B-type crystallinity is known to be strongly dependent on the processing (temperature, water content), as well as
storage conditions (relative humidity, time, temperature), which finally
influences and correlates with the stress-strain behavior of thermoplastic starch films.
Typical stress-strain curves of the starch films are shown in Fig. 4
and tensile strength, elongation at break and Young´s modulus parameters are given in Table 3. As expected, the reference starch-glycerol
film (Ref S) showed plastic behavior until fracture (Fig. 4). The mechanical properties were affected by the different film formulations and
crystallinity. In general, a high crystallinity index showed an increase in
Young´s modulus and tensile strength, which can be explained by
physical cross-links by amylose and amylopectin co-crystallization that
reduce intermolecular interactions and cause cracking. The incorporation of cellulose fibers resulted in reinforced starch films with resistance
to high stress (1086 MPa). However, these films were quite brittle,

3.5. Antioxidant activity of films by DPPH radical scavenging assay
The antioxidant extract from sunflower hulls has been described in
detail in a previous study and chlorogenic acid was identified as the
main active compound (Menzel, González-Martínez, Chiralt et al.,
2019). The calculated EC50 values using DPPH radical scavenging assay
of these films are given in Table 2 for the films with antioxidant extract
and citric acid (CaA) and cellulose fibers (CeCaA). As expected, the
phenolic compounds in the extract showed radical scavenging activity,
which was higher compared to the reference (Ref A). Citric acid significantly enhanced the antioxidant properties of the films seen as
lowering of EC50 values by a third. The exact action and migration of
citric acid in combination with the phenolic compound to protect food
against oxidation will have to be studied in more detail. However, these
films showed superior activity and, hence, are considered a valuable

source for future food contact applications.
3.6. Crystallinity and mechanical properties
The crystallinity in native granular potato starch is of the B-type
(data not shown). The diffractograms of all films are given in the Fig. 3,

Fig. 2. FESEM cross-sections (upper row) of starch films and digital images below (a – Ce films, b – Ca film, c – CeCa film, d – CaA film, e – CeCaA film).
4


Carbohydrate Polymers 250 (2020) 116828

C. Menzel

Table 2
Film thickness, internal transmittance Ti at 500 nm, moisture content, swelling in water, water vapor permeability WVP and EC50 values of starch films.
Sample

Thickness
[mm]

Ti
(500 nm)

Moisture
[g/100 g]

Swelling
[%]

WVP

[gmm/kPahm2]

EC50 values
[mg/mg DPPH]

Ce
Ca
CeCa
CaA
CeCaA
Ref S
Ref A

0.149
0.153
0.199
0.136
0.155
0.188
0.181

0.826
0.858
0.837
0.730
0.752
0.837
0.473

5.91 ± 0.31

4.66 ± 0.05
5.07 ± 0.07
4.99 ± 0.10
5.28 ± 0.14
6.03 ± 0.01
6.04 ± 0.08

327 ± 50
95 ± 14
87 ± 14
114 ± 1.4
121 ± 0.2
145 ± 11
375 ± 28

11.8 ± 1.26
10.1 ± 0.62
11.3 ± 1.03
11.0 ± 0.66
9.2 ± 0.43
9.95 ± 1.90
11.5 ± 0.54

–
–
–
89.6
88.6
–
122

Table 3
Molar mass distribution (weight-average molecular weight Mw, number-average molecular weight Mn and polydispersity D), crystallinity index Xc, Young´s modulus,
tensile strength and elongation at break of all starch films.
Sample

Ce
Ca
CeCa
CaA
CeCaA
Ref S
Ref A

Molecular weight
[x 104 g/mol]

Crystallinity index Xc

Mw

Mn

D

187 ± 7.10
0.85 ± 0.10
1.41 ± 0.26
2.56 ± 0.40
0.44 ± 0.23

357
188

102 ± 11.6
0.12 ± 0.04
0.16 ± 0.01
0.32 ± 0.06
0.29 ± 0.01
236
118

1.8
6.9
8.5
8.1
1.5
1.4
1.6

0.40
0.24
0.28
0.46
0.23
0.17
0.03

Young´s modulus

Tensile strength

Elongation at break

[MPa]

[MPa]

[%]

1086 ± 157
147 ± 21.3
145 ± 38.1
424 ± 93.7
170 ± 42.9
55 ± 20
441 ± 134

7.50 ± 2.19
3.09 ± 1.10
2.18 ± 1.45
4.83 ± 1.03
3.43 ± 0.86
4.73 ± 1.11
8.24 ± 1.54

0.82 ± 0.17
22.1 ± 8.98
16.6 ± 5.10
1.72 ± 0.93
11.2 ± 9.01

26.0 ± 5.95
4.23 ± 3.82

which can be linked to starch-cellulose interactions decreasing starch
chain mobility (Avérous, Fringant, & Moro, 2001). In addition, starch
films with incorporated antioxidant extract (CaA, CeCaA, Ref A)
showed high Young´s modulus and high tensile strength but lower extensibility. The internal microstructure might block dislocation motion
due to interactions of the phenolic compounds with the starch chains
and hence, the results in more brittle material.
The addition of citric acid (Ca, CeCa, CeCaA) had different effects on
the mechanical properties of the starch films. As expected, the addition
of citric acid decreased the extensibility but resulted in stronger films
compared to starch-glycerol films (Ref S) in terms of resistance to stress
seen as higher Young´s modulus (Fig. 4). Chemical cross-links between
starch chains and citric acid could have resulted in a reinforced network
but leads to cracking of the material. However, at the same time films
with cellulose fibers, citric acid and antioxidant extract show favorable
properties in terms of good extensibility and high tensile strength. Citric
acid has been shown to work as plasticizer and cross-linker in starch
films as well as acid hydrolyzing glycosidic bonds causing lower molecular weight, which seemed not to significantly affect the mechanical
properties (Menzel et al., 2013; Olsson, Menzel et al., 2013).

Fig. 3. SEC weight distribution of debranched starches as function of their
degree of polymerization (DP).

Fig. 4. Diffractograms of starch films.
5


Carbohydrate Polymers 250 (2020) 116828

C. Menzel

Table 4
Thermogravimetric analysis results with onset and peak temperature and weight loss.
Sample

Ce
Ca
CeCa
CaA
CeCaA
Ref S
Ref A

Onset

Peak

Weight loss [%]

[°C]

[°C]

< 150 °C

150 °C- onset

Until 350 °C

At 550 °C

276 ± 1.00
273 ± 0.41
275 ± 0.78
271 ± 1.47
273 ± 0.10
279 ± 0.32
264 ± 1.25

309 ± 1.8
308 ± 0.2
311 ± 1.2
315 ± 0.8
317 ± 0.2
305 ± 0.88
309 ± 0.51

5.2
3.2
5.0
4.6
3.0
3.5
3.5

18.3
20.8
25.6

24.3
26.1
17.6
16.8

77.9
76.4
74.5
70.3
74.5
77.9
71.2

86.3
84.7
84.2
80.0
83.8
83.6
79.3

3.7. Thermal behavior of starch films

an increase in branch-chain length of citric acid cross-linked starch has
been reported.
These films are perfect candidates as coating materials to prevent
lipid oxidation of food like nuts or cereals. In addition, the study
showed the successful utilization of sunflower hulls as agro-industrial
waste to gain added-value products in terms of active compounds and
fibers to produce green and edible food packaging.

Thermal stability of films was determined using thermal gravimetric
analysis. The weight loss as function of temperature is shown in the
supplementary (Fig. S2) with four distinct zones of weight loss. First,
temperature zone < 150 °C, which corresponds to the evaporation of
water and volatile compounds of the antioxidant extract. As expected,
starch films with antioxidant extract showed higher weight loss in that
area (Table 4). The second zone until the onset of starch degradation
(Table 4, weight loss 150 °C until onset at ∼270 °C) represents the
weight loss due to the degradation of glycerol and citric acid, which
decompose above 148 °C (Barbooti & Al-Sammerrai, 1986). All starch
films with citric acid showed higher weight loss up to ∼270 °C. The
third degradation zone from onset temperature to 350 °C is typical for
the decomposition of starch. The fourth zone above 350 °C, represents
the residual weight of starch films after decomposition, which differed
due to the higher ash content from the antioxidant extract (Menzel,
González-Martínez, Chiralt et al., 2019). The thermal properties of
starch films were not negatively affected by the addition of citric acid.

Author contribution
The author, Carolin Menzel, wrote the project, which was financially covered by the Swedish research council by a personal grant to
the same person. All analysis and interpretation including figures and
tables were carried out by the author. The manuscript was entirely
written and revised by the author.
Acknowledgement
This work was supported by the Swedish Research Council Formas
[2015-00550].

3.8. Barrier properties and swelling

Appendix A. Supplementary data

Changes in water vapor permeability of all films were not significantly affected and were between 9.2 and 11.8 gmm/kPahm2,
which is expected for starch-glycerol films (Menzel, González-Martínez,
Chiralt et al., 2019). Previous reports showed that water vapor permeability was improved in solution cast starch films with added citric
acid (Olsson, Hedenqvist, Johansson, & Järnström, 2013). However, the
film processing by melt-blending and compression molding did not
improve film properties in the same way, which has to be studied in
further detail.
The swelling of starch film in water is given in Table 2 and calculated as weight increase. The glycerol-starch films were able to swell
slightly (145 %). The incorporation of citric acid in all film formulations
showed a decrease in swelling, which indicates a higher integrity of the
films through ester bond and cross-links between starch. The increase in
swelling of films with cellulose fibers was expected due to the swelling
capability of the fibers. However, the reference film containing only
antioxidant extract and glycerol showed largest degree of swelling (375
%), which was unexpected and has to the best of our knowledge not
been reported yet.

Supplementary material related to this article can be found, in the
online version, at doi: />References
Avérous, L., Fringant, C., & Moro, L. (2001). Plasticized starch–cellulose interactions in
polysaccharide composites. Polymer, 42(15), 6565–6572.
Barbooti, M. M., & Al-Sammerrai, D. A. (1986). Thermal decomposition of citric acid.
Thermochimica Acta, 98, 119–126.
Benito-González, I., López-Rubio, A., & Martínez-Sanz, M. (2019). High-performance
starch biocomposites with celullose from waste biomass: Film properties and retrogradation behaviour. Carbohydrate Polymers, 216, 180–188.
Castaño, J., Rodríguez-Llamazares, S., Sepúlveda, E., Giraldo, D., Bouza, R., & Pozo, C.
(2017). Morphological and structural changes of starch during processing by melt
blending. Starch - Stärke, 69(9–10), Article 1600247.

Chiralt, A., Menzel, C., Hernandez-García, E., Collazo, S., & Gonzalez-Martinez, C. (2020).
Chapter 6 - Use of by-products in edible coatings and biodegradable packaging materials for food preservation. In N. Betoret, & E. Betoret (Eds.). Sustainability of the
food system (pp. 101–127). Academic Press.
Economic, U.N.D.o, & Affairs, S (2018). The sustainable development goals report 2018.
Galanakis, C. M. (2012). Recovery of high added-value components from food wastes:
Conventional, emerging technologies and commercialized applications. Trends in
Food Science & Technology, 26(2), 68–87.
Gennadios, A., Weller, C. L., & Gooding, C. H. (1994). Measurement errors in water vapor
permeability of highly permeable, hydrophilic edible films. Journal of Food
Engineering, 21(4), 395–409.
Hulleman, S. H. D., Kalisvaart, M. G., Janssen, F. H. P., Feil, H., & Vliegenthart, J. F. G.
(1999). Origins of B-type crystallinity in glycerol-plasticised, compression-moulded
potato starches. Carbohydrate Polymers, 39(4), 351–360.
Hutchings, J. B. (1999). Instrumental specification. Food colour and appearance.
Springer199–237.
Jambeck, J. R., Geyer, R., Wilcox, C., Siegler, T. R., Perryman, M., Andrady, A., & Law, K.
L. (2015). Plastic waste inputs from land into the ocean. Science, 347(6223),
768–771.
Luchese, C. L., Uranga, J., Spada, J. C., Tessaro, I. C., & de la Caba, K. (2018). Valorisation
of blueberry waste and use of compression to manufacture sustainable starch films
with enhanced properties. International Journal of Biological Macromolecules, 115,

4. Conclusions
4.1. Three principle approach – crosslinking, antioxidant and reinforcement
The combination of citric acid as cross-linker, cellulose fibers as
reinforcement and active extract as antioxidant resulted in films with
superior properties. Films had inherent antioxidant activity towards
DPPH radicals. The inclusion of cellulose fibers improved starch performance in terms of resistance towards stress and at the same time
keeping sufficient extensibility. Furthermore, this is the first time that
6

Carbohydrate Polymers 250 (2020) 116828

C. Menzel
955–960.
Menzel, C., Olsson, E., Plivelic, T. S., Andersson, R., Johansson, C., Kuktaite, R., & Koch,
K. (2013). Molecular structure of citric acid cross-linked starch films. Carbohydrate
Polymers, 96(1), 270–276.
Menzel, C., González-Martínez, C., Chiralt, A., & Vilaplana, F. (2019). Antioxidant starch
films containing sunflower hull extracts. Carbohydrate Polymers, 214, 142–151.
Menzel, C., González-Martínez, C., Vilaplana, F., Diretto, G., & Chiralt, A. (2019).
Incorporation of natural antioxidants from rice straw into renewable starch films.
International Journal of Biological Macromolecules.
Müller, C. M. O., Laurindo, J. B., & Yamashita, F. (2009). Effect of cellulose fibers addition
on the mechanical properties and water vapor barrier of starch-based films. Food
Hydrocolloids, 23(5), 1328–1333.
Olsson, E., Hedenqvist, M. S., Johansson, C., & Järnström, L. (2013). Influence of citric
acid and curing on moisture sorption, diffusion and permeability of starch films.
Carbohydrate Polymers, 94(2), 765–772.
Olsson, E., Menzel, C., Johansson, C., Andersson, R., Koch, K., & Järnström, L. (2013). The
effect of pH on hydrolysis, cross-linking and barrier properties of starch barriers
containing citric acid. Carbohydrate Polymers, 98(2), 1505–1513.
Pérez, S., & Bertoft, E. (2010). The molecular structures of starch components and their
contribution to the architecture of starch granules: A comprehensive review. Starch -

Stärke, 62(8), 389–420.
Seligra, P. G., Medina Jaramillo, C., Famá, L., & Goyanes, S. (2016). Biodegradable and
non-retrogradable eco-films based on starch–glycerol with citric acid as crosslinking
agent. Carbohydrate Polymers, 138, 66–74.

Shi, R., Zhang, Z., Liu, Q., Han, Y., Zhang, L., Chen, D., ... Tian, W. (2007).
Characterization of citric acid/glycerol co-plasticized thermoplastic starch prepared
by melt blending. Carbohydrate Polymers, 69(4), 748–755.
UNenvironmentprogramme (2019). Plastic recycling: An underperforming sector ripe for a
remake.
Valdés, A., Mellinas, A. C., Ramos, M., Garrigós, M. C., & Jiménez, A. (2014). Natural
additives and agricultural wastes in biopolymer formulations for food packaging.
Frontiers in Chemistry, 2, 6.
van Soest, J. J. G., Hulleman, S. H. D., de Wit, D., & Vliegenthart, J. F. G. (1996). Changes
in the mechanical properties of thermoplastic potato starch in relation with changes
in B-type crystallinity. Carbohydrate Polymers, 29(3), 225–232.
Vilaplana, F., & Gilbert, R. G. (2010). Two-dimensional size/branch length distributions
of a branched polymer. Macromolecules, 43(17), 7321–7329.
Wilpiszewska, K., & Czech, Z. (2014). Citric acid modified potato starch films containing
microcrystalline cellulose reinforcement – properties and application. Starch - Stärke,
66(7–8), 660–667.

7