Using gelatin protein to facilitate paper thermoformability

Full text

Using gelatin protein to facilitate paper thermoformability
Alexey Khakalo
a
, Ilari Filpponen
a,
⇑
, Leena-Sisko Johansson
a
, Alexey Vishtal
b
, Arcot R. Lokanathan
a
,
Orlando J. Rojas
a,c
, Janne Laine
a
a
Aalto University School of Chemical Technology, Department of Forest Products Technology, P.O. Box 16300, 00076 Aalto, Finland
b
VTT Technical Research Centre of Finland, P.O. Box 1603, Koivurannantie 1, Jyväskylä 40101, Finland
c
North Carolina State University, Departments of Forest Biomaterials and Chemical and Biomolecular Engineering, Raleigh, NC 27695, USA
article info
Article history:
Received 7 July 2014
Received in revised form 23 September 2014
Accepted 24 September 2014
Available online 2 October 2014
Keywords:
Cellulose modification
Gelatin
Paper formability
Paper extensibility
Packaging
abstract
One of the main challenges of fiber-based packaging materials is the relatively poor elongation of
cellulose under stress, which limits formability and molding in related products. Therefore, in this inves-
tigation we first used cellulose thin films and surface sensitive tools such as quartz crystal microbalance
(QCM-D), surface plasmon resonance (SPR) and X-ray photoelectron spectroscopy (XPS) to evaluate the
cellulose–gelatin interactions. It was found that the highest adsorption of gelatin onto cellulose occurred
at the isoelectric pH of the protein. Based on this and other results, a gelatin loading is proposed to
facilitate molecular and surface interactions and, thus to improve the formability of cellulose-based
materials in paper molding. Aqueous gelatin solutions were sprayed on the surface of wet webs com-
posed of softwood fibers and the chemical and mechanical changes that occurred were quantified. Upon
gelatin treatment the elongation and tensile strength of paper under unrestrained drying was increased
by 50% (from 10% to 14%) and by 30% (from 59 to 78 N m/g), respectively. The mechanical perfor-
mance of gelatin-treated fibers was further improved by glutaraldehyde-assisted cross-linking. The pro-
posed approach represents an inexpensive and facile method to improve the plasticity of fiber networks,
which otherwise cannot be processed in the production of packaging materials by direct thermoforming.
Ó2014 Elsevier B.V. All rights reserved.
1. Introduction
Environmental concerns and a more reasonable use of fossil
resources have amounted to a growing interest for renewable
and biodegradable packaging materials [1]. Cellulose is one of the
most abundant biopolymers in the biosphere and the main raw
material in production of paper and board [2]. However, one of
the main challenges of using cellulose-based packaging materials
(such as cups, plates, trays and food containers) is the poor
formability of cellulose [3]. Moreover, the barrier properties and
moisture resistance of cellulose-based materials are not competi-
tive when compared against traditional plastics. While barrier
properties and moisture resistance can be improved by introducing
additional coating layers, paperboard formability requires mechan-
ical and chemical modifications of fibers and the fiber network
structure to make it feasible.
The term ‘‘formability’’ describes the ability of a material to
undergo plastic deformation without damage and it is an essential
property for 3D-forming processes, where various advanced
shapes can be produced. As of today, the most stretchable bio-
based composite materials have been prepared by reinforcing duc-
tile hydrophobic matrices with natural fibers [4]. However, the
interfacial adhesion between the components is not very good
and chemical modification of natural fibers is typically required
to improve the mechanical properties of such composites [5,6].It
is also important to note here that natural fibers are typically used
to a lesser extent when compared to the hydrophobic component
of the composites.
The main features of paper and paperboard for deep-drawing
processes have been recently studied and the ability of a material
to deform under applied stress without a failure has been high-
lighted [3]. The combination of high-consistency wing defibrator
refining and subsequent low-consistency valley beating of fibers
(laboratory process only) was proposed as a mechanical treatment
to improve the extensibility and elongation at break without dete-
riorating the strength properties of resulting paperboard [7]. High-
consistency treatment decreased the axial stiffness of fibers while
the low-consistency refining improved fiber–fiber bonding by
strengthening the fibers. It was also reported that the improved
bonding between fibers promotes the stretch potential of curly
fibers making the paper more extensible [8]. Furthermore, the
http://dx.doi.org/10.1016/j.reactfunctpolym.2014.09.024
1381-5148/Ó2014 Elsevier B.V. All rights reserved.
⇑
Corresponding author. Tel.: +358 503605623.
E-mail address: erkko.filpponen@aalto.fi (I. Filpponen).
Reactive & Functional Polymers 85 (2014) 175–184
Contents lists available at ScienceDirect
Reactive & Functional Polymers
journal homepage: www.elsevier.com/locate/react

importance of paper drying was also acknowledged, i.e., the
shrinkage of paper upon unrestrained drying (negative strain)
could be further recovered to increase the overall extensibility of
a paper.
The extensibility can also be improved by modifying the fiber or
fiber network structure. For example, hydroxypropylation of cellu-
lose was used to produce highly extensible and translucent paper
[9]. Stretchable paper-based material has been prepared by con-
verting a selectively dissolved fiber surface into the matrix
[10,11]. Moreover, the elastic properties of a paper have been
improved by applying film forming methylcellulose and lignosulfo-
nate as surface sizing agents [12]. Approximately 50% increase in
strain to failure was observed by employing a recombinant
cellulose crosslinking protein [13]. Agar has also been used as an
additive for improving the extensibility of paper [14]. In addition,
the combination of gelatin and nanocellulose substrates has been
shown to produce materials with excellent stress performance
[15–17].
Gelatin is a mixture of proteins obtained by the hydrolysis of
collagen, the most abundant protein in waste products like skins,
connective tissues, bones and cartilage of predominant bovine
animals. When heated to 40 °C, gelatin dissolves in water with a
formation of random coiled chains. Upon cooling, gelatin chains
partially recover the original triple-helix structure of collagen.
Thus, gelatin gels form an ensemble of physically interconnected
triple helices, which are held together by intermolecular hydrogen
bonds. From the point of view of its chemical structure, gelatin is a
weak polyampholyte. Typically, approximately 13% of gelatin
backbone is positively charged (lysine and arginine), 12% is neg-
atively charged (glutamic and aspartic acid) and 11% is hydro-
phobic in nature (leucine, isoleucine, methionine and valine)
[18]. Gelatin is widely utilized in the food industry. In pharmaceu-
tical field gelatin is used in controlled delivery systems [19,20] and
for fabrication of three-dimensional gelatin-based polymer
scaffolds for tissue-engineering and wound dressing purposes
[21–23]. Moreover, gelatin-based films and interpenetrating
polymeric networks have been proposed as alternatives to synthetic
food packaging materials [2,16,24].
Apart from the aforementioned advantages of gelatin, there are
some drawbacks that hinder its applications, among which poor
mechanical properties and moisture sensitivity are the most limit-
ing ones. Different approaches to overcome these drawbacks have
been proposed but chemical cross-linking is by far the most effec-
tive method to improve mechanical, thermal, and water-resistance
properties of gelatin. A variety of chemicals, biomolecules and
enzymes have been used for cross-linking, such as glutaraldehyde
[18,19], epoxides [25], glyoxal [26], isocyanates [27], formaldehyde
[25,26], glyceraldehyde [28], genipin [29] and transglutaminase
[30]. However, glutaraldehyde is the most widely used due to its
commercial availability, short-reaction time and low cost. In addi-
tion, it has an excellent ability to stabilize collagenous materials
and therefore to improve the strength properties and water resis-
tance [18].
Gelatin has been widely studied for its gel and film forming
abilities in combination with cross-linker, like glutaraldehyde, or
with plasticizer, like glycerol and sorbitol. However, despite the
growing interest, to our knowledge there are no reports available
on the application of gelatin for production of highly extensible
paper and paperboard. In this communication, we first unveil the
nature of cellulose–gelatin interactions by using quartz crystal
microbalance with dissipation monitoring (QCM-D) and surface
plasmon resonance (SPR) to then apply aqueous gelatin solution
via spraying on cellulose fiber networks. As a result, an improved
formability of paper was achieved. The morphology and composi-
tion of adsorbed gelatin layer were analyzed by scanning electron
microscopy (SEM), atomic force microscopy (AFM) and X-ray
photoelectron spectroscopy (XPS), all of which provide supporting
evidence of the mechanism and mode of action of gelatin as an
inexpensive and facile method to endow fiber networks with plas-
ticity to make them suitable for 3D thermal forming.
2. Experimental
2.1. Materials
First thinning bleached softwood kraft fiber sheets (cellulose
81.7%, xylan 9.2%, glucomannan 9.0% and total lignin <0.5%) were
provided by Pietarsaari mill, UPM-Kymmene. Gelatin from porcine
skin (Type A, 300 g Bloom gel strength, #232-554-6), glutaralde-
hyde solution (50 wt.% in H
2
O, #340855) and polystyrene (280 kDa
molecular weight, #182427) were obtained from Sigma–Aldrich
(US). Trimethylsilyl cellulose (TMSC) was synthetized as described
elsewhere [31].
2.2. Preparation of cellulose surfaces
Cellulose-gelatin interactions were investigated by using gold-
coated sensors in SPR and QCM-D experiments. The sensors (SPR
gold chips or QCM crystals) were first cleaned with UV/ozone
treatment for 15 min followed by spin coating with 0.1 wt.% poly-
styrene in toluene (4000 rpm, 60 s). Prepared polystyrene-coated
sensors were then dried in an oven at 60 °C for 10 min to ensure
a uniform hydrophobic layer suitable for trimethylsilyl cellulose
(TMSC) deposition. TMSC was deposited on the polystyrene-coated
sensors by using the Langmuir–Schaeffer (LS) lifting deposition
technique as described by Tammelin et al. [32]. The TMSC layer
was then converted to cellulose via desilylation with hydrochloric
acid vapor as described elsewhere [33]. The crystallinity degree,
thickness, and roughness of the LS-cellulose films prepared in the
same manner have previously been observed to be 54%, 17.8 nm,
and 0.5 nm, respectively [34]. Before QCM-D and SPR experiments
the cellulose films were allowed to stabilize overnight in the
appropriate buffer solution.
2.3. Adsorption experiments using cellulose sensors
Prior to QCM-D and SPR adsorption experiments, gelatin was
dissolved in Milli-Q water and dialyzed using a 10–12 kDa mesh
membrane tube (SpectraPor, Spectrumlabs) and freeze-dried.
Isoelectric point of gelatin was measured using zeta-potential ana-
lyzer (Malvern Zetasizer Nano ZS device, Malvern Instruments,
Malvern, UK). Dialyzed gelatin was dissolved in 10 mM acetate,
phosphate and bicarbonate buffers at pH 4, 5.8 and 10, respectively
at 45 °C for 30 min to yield 0.1 mg/mL concentration. Afterwards,
gelatin solution was filtered using 0.45
l
m filters and degassed.
Both QCM-D and SPR experiments were performed at a constant
flow rate of 100
l
L/min and the temperature was maintained at
21 °C until adsorption plateau was reached. Thereafter, rinsing
with polymer-free buffer solution was applied to ascertain the irre-
versible binding of gelatin to cellulose surfaces pre-adsorbed on
the (QCM/SPR) sensors. Finally, the sensors were washed with
Milli-Q water and stored in desiccator until further investigation.
Each set of experiments was performed at least two times.
2.4. Adsorbed mass by using quartz crystal microbalance and surface
plasmon resonance
Gelatin adsorption onto cellulose and properties of adsorbed
gelatin layer were investigated by using quartz crystal microbal-
ance with dissipation (QCM-D) monitoring (E4 instrument, Q-
Sense AB, Sweden) and surface plasmon resonance (SPR) unit
176 A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184

(SPR Model Navi 200, Oy BioNavis Ltd., Tampere, Finland). With
QCM-D, changes in mass and film viscoelastic properties on the
sensor surface are detected by simultaneous monitoring changes
in frequency (
D
f) and dissipation (
D
D), respectively, as a function
of time at the fundamental resonance frequency (5 MHz) and its
overtones (15, 25, 35, 45, 55, and 75 MHz). The principles of
QCM-D operation and data analysis can be found elsewhere
[35,36]. If the adsorbed layer is rigid, uniformly distributed on
the surface, and small compared to the mass of the solid support
(crystal), the Sauerbrey’s relation can be applied to calculate the
mass change upon adsorption (
D
m)[37]:
D
m¼c
D
f
nð1Þ
where C= 17.7 ng/cm
2
for 5 MHz crystal,
D
fis change in frequency,
and nis the overtone number. However, if the adsorbed layer is soft
or highly hydrated, the Voigt viscoelastic model (calculations by Q-
Tools software, version 2.1, Q-Sense, Västra Frölunda, Sweden) is
applicable. This model utilizes frequency and dissipation data from
several overtones and can be represented as a system consisting of a
spring and a dashpot filled with viscous fluid connected in parallel
[38]. The density of the adsorbed gelatin layer in this latter case was
assumed to be 1200 g/m
3
. Frequency shifts (
D
f) at the third over-
tone (15 MHz) are presented since this overtone usually has the
best signal-to-noise ratio.
In SPR experiments the so-called SPR angle can be used to
derive the thickness of adsorbed layer: [39]
d¼l
d
2
D
SPR angle
mðn
a
n
0
Þð2Þ
where
D
SPR angle
is a change in the SPR angle, ld is a characteristic
evanescent electromagnetic field decay length (240 nm), estimated
as 0.37 of the light wavelength; mis a sensitivity factor for the sen-
sor (109.94°/RIU) obtained after calibration of the SPR, n
0
is the
refractive index of the bulk solution (1.33 RIU) and n
a
is the refrac-
tive index of the adsorbed layer (assumed value of 1.60 RIU mea-
sured for crystalline proteins).
The surface excess concentration was calculated according to
Eq. (3):
M¼d
v
ð3Þ
where dis the calculated thickness of the adsorbed layer and
m
is
the specific density of the layer (1.35 g/cm
3
)[40].
2.5. Surface topography by atomic force microscopy (AFM)
Topographical changes on the cellulose surfaces after gelatin
adsorption were analyzed by AFM equipped with Nanoscope IIIa
Multimode scanning probe microscope from Digital Instruments
Inc. (Santa Barbara, CA, USA). 1 1
l
m
2
images were recorded by
using silicon cantilevers in air scanned via the tapping mode. Prior
to measurements, the samples were allowed to dry in desiccator at
room temperature at least overnight. At least three different
regions on each sample were analyzed. Image analyses were per-
formed using NanoScope Analysis 1.2 software, no image process-
ing except flattening was done. The rms surface roughness was
measured from 1 1
l
m
2
scan sizes.
2.6. Gelatin-modified paper
Gelatin-modified paper was prepared by repeated spraying of
aqueous 4 wt.% gelatin solution (with respect to dry cellulose
fibers) on freshly prepared cellulosic fiber handsheets before wet
pressing, as described in Ref. [41]. Samples with different gelatin
loading were obtained. Prior to handsheet preparation the fibers
were pre-treated combining high (Wing defibrator) and low
(Valley beater) consistency refining until the degree of refining
measured by Schopper–Riegler (SR) was 25 [7]. Next, the fibers
were washed to Na-form in order to remove adsorbed metal ions
and water-soluble substances, according to a procedure described
by Laine et al. [42]. Handsheets were prepared using deionized
water according to ISO 5269-1 except that a grammage of
200 g/m
2
was used. Gelatin was swelled in deionized water for
1 h and then dissolved at 45 °C for 30 min. Gelatin solution
(4 wt.%) was used for spraying by using a universal spray gun
(Wagner W 140P, J. Wagner GmbH, Germany). For the preparation
of gelatin-treated paper containing cross-linker, glutaraldehyde
solution was sprayed on top of the sprayed gelatin layer. Finally,
samples were air-dried (23 °C and 50% RH) between two plastic
wire nets, which enabled the handsheets to shrink freely without
excessive curling. All the characterizations were conducted for
the freely dried samples, unless otherwise stated.
2.7. Tensile testing
The tensile strength and elongation at break of the gelatin-
treated papers were measured with a MTS 400/M (MTS Systems,
USA) vertical tensile tester with a load cell of 2 kN equipped with
TestWorks 4.02 measuring program and run according to
SCAN-P38. At least 20 replicate specimens from each gelatin-
treated paper type were measured.
2.8. Sheet formability
Formability strain and strength were measured using a 2D
formability tester developed by VTT. This unit is equipped with a
double-curved heated press (temperatures up to 250 °C), blank
holders and an IR sensor (Omega
Ò
OS36) to measure the tempera-
ture of the paper. Typically, a paper with a grammage range from
80 to 250 g/m
2
can be preheated to the die temperature within
0.5–0.7 s. In practise this means that the temperature of the paper
at the moment of forming is close to that of the die. Fig. 1 presents
the basic operation principle of 2D formability tester and a photo-
graphic image of the formed paper strip.
The forming proceeds as follows: a paperboard sample (15 mm
wide 65 mm long) is fixed by the two blank holders. The press is
then moved into contact with the sample and retained still for 0.5 s
in order to preheat the sample. Then, press continues a downward
movement until breakage of the sample. The velocity of the form-
ing press was 10 mm/s [43]. The formability strain of the samples
was measured as an average value collected from 10 samples at die
temperatures of 23 (room temperature), 45, 60, 75, 90, 105, 120
Fig. 1. Schematic representation of 2D formability tester and formed strip of paper.
A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184 177

and 135 °C. In order to evaluate influence of the moisture content
on the paper formability, the paper samples were conditioned at
two different relative humidity levels 50% and 75% (23 °C) prior
to forming.
2.9. Fiber chemical changes by Fourier transformed infrared
spectroscopy (FTIR)
Bulk chemical characteristics and bonding of both cross-linked
and non-cross-linked gelatin-treated paper, as well as reference
samples were investigated by an Avatar 360 FTIR spectrometer
(Thermo Nicolet). Spectra were recorded at room temperature in
absorbance mode within the wavenumber range of 400–
4000 cm
1
at 4 cm
1
resolution using KBr pellets. Each set of data
acquisition was performed at least two times, for each spectrum
the number of scans was set to 32 and CO
2
and H
2
O corrections
were carried out prior to every measurement. Omnic 6.0a software
was used for data acquisition and analysis.
2.10. Imaging by scanning electron microscopy (SEM)
A FEI Quanta 200 scanning electron microscope (SEM) was uti-
lized for in plane and cross section imaging. The SEM was operated
in back scattered electron mode at an emission current of roughly
100
l
A, and using an accelerating voltage of 12.5 kV or 15 kV. The
working distance was set to 10 mm and a spot size of 5–6 was
used. The samples were previously coated with evaporated carbon
using a BALZERS SCD 050 sputter coater equipped with a non-
rotating base. The cross sections were prepared by a Hitachi
IM4000 cross section cutter prior to the SEM investigation.
2.11. Surface chemical composition via X-ray photoelectron
spectroscopy (XPS)
Surface chemical composition of fibers as well as QCM surfaces
was investigated with XPS Kratos Axis Ultra instrument (Kratos
Ltd., Manchester, UK). The samples were measured after an over-
night pre-evacuation using monochromated Al K
a
irradiation at
100 W (8 mA, 12.5 kV) under neutralization. Wide scan spectra
were measured using 1 eV step and 80 eV pass energy, and the
high resolution measurements were recorded with 0.1 eV step
and 20 eV pass energy. Experimental conditions were monitored
with an in-situ reference (100% cellulose filter paper). All spectra
were collected at an electron takeoff angle of 90°. Each sample
was analyzed on at least 3 different locations and the mean of
three values are presented. The area and depth of analysis was
1mm
2
and less than 10 nm, respectively. No sample degradation
due to ultrahigh vacuum or X-ray radiation was observed during
the XPS measurements. Both elemental wide-region data and high
resolution spectra of carbon (C 1s) and oxygen (O 1s) were col-
lected. The relative amounts of carbon, oxygen, nitrogen (protein
marker), and silicon were determined from C 1s, O 1s, N 1s, and
Si (Si 2p or Si 2s or 1s) signals from low-resolution scans, and
the high-resolution C1s spectra were curve fitted for further chem-
ical analysis, using parameters defined for cellulosic materials [44].
2.12. Stamping of gelatin-treated paper
Stamping of prepared gelatin-treated papers was conducted at
the laboratory scale using 3D forming device (Stora Enso RC, Ima-
tra). The forming conditions were as follows: temperature of the
cavity 140 °C, speed of the die 50 mm/s, blank holding force 3 kN,
moisture content of paper ca. 8% (conditioned at 50% RH and
23 °C).
3. Results and discussions
3.1. Adsorption of gelatin on cellulose
Cellulose–gelatin interactions were studied by using Langmuir–
Schaefer (LS) cellulose films [32]. The LS cellulose films are made
up of pure cellulose and have been taken as a good surface chem-
istry model of cellulose fibers. LS films are typically very smooth
and the surface quality can be tuned because the molecular trans-
fer ratio (and resulting degree of molecular packing) can be regu-
lated during the deposition stage. Moreover, the absence of voids
and lower surface roughness of LS films simplify the interpretation
of data from adsorption and interfacial interactions. Interactions
between slightly negatively charged cellulose LS films and gelatin
were studied with QCM-D and SPR under identical experimental
conditions. It should be noted here that the electrostatic charge
of gelatin is dependent on the solution pH, i.e., gelatin is negatively
charged above its isoelectric pH of 5.8 and positively charged
below it. Gelatin adsorption on cellulose LS films at pH 4, 5.8 and
10 is presented in Fig. 2.
It was found that the gelatin adsorption is rather irreversible as
no significant desorption was observed after rinsing with the back-
ground buffer solution. Moreover, the highest adsorbed amount
was detected at pH 5.8 which is the isoelectric pH of gelatin. This
is in line with the well-known fact that protein adsorption is max-
imized near the isoelectric pH [45]. It has also been speculated that
the gelatin adsorption is enhanced by its flexible configuration at
the interface so that the surface of cellulose can accommodate a
large amount of protein molecules, i.e., by suitably changing their
structural orientation at the polysaccharide interface [46]. Similar
observations have been reported for the gelatin adsorption on non-
cellulosic hydrophobic surfaces such as phosphatidylcholine coated
silica [47]. At pH 4 gelatin is positively charged and it behaves as a
polyelectrolyte (pKa of glutamic COOH is 4.3). Therefore, it was
expected for the adsorption of gelatin onto slightly negatively
charged cellulose to be high considering the electrostatic interac-
tions. However, gelatin adsorption at pH 4 was found to be less than
at pH 10, when gelatin is negatively charged (pKa of arginine is 9.0).
The amount of free amino groups in gelatin molecules is much
higher than carboxylic acid groups, therefore, if compared to obser-
vations at pH 4, there are more unprotonated
e
-amino groups than
protonated carboxylic acid groups at pH 10 [18]. Overall, there is
indication that hydrogen bonding and other non-specific interac-
tions are important in cellulose-gelatin interactions.
The thicknesses of the adsorbed gelatin layers determined by
QCM-D were found to be significantly higher than those obtained
from SPR experiments. This confirms the hydrogel-like nature of
gelatin in aqueous solution. In fact, the amount of coupled water
in the adsorbed protein can be calculated by comparing the
QCM-D and SPR data since difference in the calculated adsorbed
mass corresponds to such contribution. The water content of gela-
tin adsorbed at the isoelectric pH is slightly lower (92%) when
compared to those at pH 4 and 10 (96% in both cases). This is
likely attributed to the reduced amount of charged groups in gela-
tin at the isoelectric pH, which decreases water coupling or hydra-
tion. The respective adsorbed masses for gelatin were calculated by
applying Voigt viscoelastic model and SPR mass (Eqs. (2) and (3))
since the adsorbed gelatin layers did not meet the Sauerbrey’s con-
ditions (see Table 1).
The LS cellulose films with adsorbed gelatin were further inves-
tigated with X-ray photoelectron spectroscopy (XPS). Fig. 2c
includes the oxygen/nitrogen atom % ratios (O/N) of respective sur-
faces. As expected, the highest amount of nitrogen (lowest O/N
ratio) was found from the surface in which the gelatin was
adsorbed at pH 5.8. XPS results were in good agreement with the
adsorption studies (QCM-D and SPR).
178 A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184

Finally, topographical changes of gelatin-modified LS cellulose
surfaces were studied with atomic force microscopy (AFM). At
pH 5.8, the adsorbed gelatin layer displays globular conformation,
whereas at pH 4 and pH 10 the layer appears to contain more
prominent, elongated structures (Fig. 3). In addition, a higher
roughness is found for the samples obtained after adsorption at
pH 4 and pH 10 (0.764 and 0.817, respectively) compared to that
found at pH 5.8 (0.599). These observations can be ascribed to
the reduced solution stability of gelatin molecules at isoelectric
pH while the large degree of gelatin swelling at pH 4 and 10 is a
contributing factor to the increased surface roughness under these
conditions.
Based on the QCM and SPR results, it is hypothesized that the
addition of gelatin, its effect on hydrogen bonding and interac-
tion via amino acids and hydrophobic effects, could be contribut-
ing factors to achieve enhanced fiber extensibility. Therefore, the
following sections summarize the effect of gelatin when applied
on the surface of fibers, under optimal conditions to maximize
adsorption.
3.2. Mechanical performance of gelatin-modified paper
Gelatin-modified paper was obtained by repeated spraying of
aqueous 4 wt.% gelatin solution (with respect to dry cellulose
fibers) on freshly prepared handsheets before wet pressing. Sam-
ples with different gelatin loading were obtained. Typical stress–
strain curves of the gelatin-modified paper are presented in
Fig. 4a and a summary of the key mechanical properties are
reported in Table 2.
Gelatin upon drying tends to recover partial collagen-like triple-
helix structure stabilized mainly by the formation of inter-chain
hydrogen bonds between carbonyl and amine groups. Thus, the
films containing only gelatin are brittle and susceptible to crack
due to the strong cohesive energy density of the polymer [24].
However, the application of gelatin as a minor component in a cel-
lulosic composite may contribute to the fiber–fiber bonding by fill-
ing the voids and by increasing the contact areas between the
fibers. As evident from Table 2, tensile strength of the gelatin-mod-
ified paper increased until 8 wt.% gelatin content, indicating that
Fig. 2. Adsorption of gelatin on LS cellulose films measured as a shift in QCM frequency (third overtone,
D
f) (a), by SPR (reported as change in SPR angle) and resulting
changes in surface chemical composition from XPS data (c) as indicated by the ratios of oxygen and nitrogen atom concentration on the surface of the cellulose surfaces after
gelatin adsorption at three different pH values. The O/N ratio of gelatin is given as a reference.
Table 1
Calculated QCM-D and SPR adsorbed masses (
D
m, mg/m
2
), layer thicknesses (nm), and coupled water of gelatin adsorbed from aqueous solutions on LS cellulose surfaces. The
QCM adsorbed mass was calculated by using the Voigt viscoelastic model.
QCM-D (wet weight) SPR (dry weight) Water (%)
Thickness (nm)
D
m(mg/m
2
) Thickness (nm)
D
m(mg/m
2
)
pH 4 7.75 ± 0.05 9.3 ± 0.2 0.28 ± 0.05 0.38 ± 0.05 96
pH 5.8 21.64 ± 0.14 26 ± 0.8 1.68 ± 0.1 2.27 ± 0.1 91
pH 10 12.77 ± 0.1 15.3 ± 0.4 0.48 ± 0.05 0.65 ± 0.05 96
A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184 179

an optimal cellulose–gelatin interaction is reached at this particu-
lar loading level. Moreover, the density of 8 wt.% gelatin-contain-
ing paper increased from 740 (reference sample) to 820 kg/m
3
,
which is still in the range of typical densities of commercial paper-
boards utilized for packaging purposes [48].
The strength properties of the treated paper slightly decreased
at gelatin loadings higher than 8 wt.%. However, the extensibility
(strain to failure) and TEA Index (the tensile energy absorbed by
the sample before failure) of the gelatin-modified paper is found
to increase steadily with gelatin within the whole range of loading.
Fig. 3. AFM height images and corresponding roughness profiles of unmodified cellulose (a), and cellulose with adsorbed gelatin at pH 4 (b), 5.8 (c) and 10 (d). The Z-range of
all the images is 5 nm.
Fig. 4. Stress–strain curves of gelatin-modified paper (a) and glutaraldehyde cross-linked gelatin-modified paper (b).
Table 2
Mechanical properties of gelatin-modified paper and glutaraldehyde cross-linked gelatin-modified paper.
Sample Density (kg/m
3
) Strain to failure (%) Tensile index (N m/g) Stiffness index (kNm/g) TEA index (J/g)
Gelatin addition (wt.%)
0 740 9.7 ± 0.5 59.3 ± 3.1 2.6 ± 0.3 3.3 ± 0.2
4 770 11.4 ± 0.4 70.3 ± 4.4 3.0 ± 0.4 4.8 ± 0.4
8 820 14.3 ± 0.6 77.9 ± 5.6 3.4 ± 0.3 6.3 ± 0.5
12 760 18.0 ± 1.2 74.2 ± 4.2 2.7 ± 0.4 7.7 ± 0.7
16 680 21.1 ± 1.4 73.2 ± 5.8 2.5 ± 0.3 9.3 ± 0.7
20 710 21.6 ± 2.0 70.2 ± 5.3 2.3 ± 0.4 9.3 ± 0.8
4 wt.% gelatin addition + glutaraldehyde
4 + 0.5 730 14.1 ± 0.6 77.0 ± 5.9 3.0 ± 0.4 6.4 ± 0.3
4 + 1.0 740 15.0 ± 0.8 76.8 ± 5.8 3.0 ± 0.4 6.7 ± 0.5
4 + 1.5 730 14.3 ± 0.7 73.3 ± 5.7 3.0 ± 0.4 6.2 ± 0.5
180 A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184

Fig. 5. Formability strain (a and c) and strength (b and d) as a function of processing temperature of paper as well as gelatin-treated paper with and without crosslinking.
Samples were conditioned at relative humidity levels of 50% (a–b) and 75% (c–d).
x
denotes the water content of the samples.
Fig. 6. SEM micrographs of reference sample (a and d), gelatin-treated paper (4 wt.% gelatin) without (b and e) and with crosslinking (1 wt.% of glutaraldehyde) (c and f).
Plane view images are shown in panels a–c and the respective cross-section are included in panels d–f.
A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184 181

It is important to note that gelatin application facilitates drying
shrinkage of the treated paper. This was observed by comparing
the extensibilities of samples that were dried under restraint. For
instance, while dried under restraint, the extensibility of the sam-
ple treated with 4% gelatin (GEL 4%) was 2% lower than that of the
sample subjected to free drying. Furthermore, the shrinkage was
more pronounced for the samples with higher gelatin content
(data not shown). Therefore, it is reasonable to postulate that the
increased extensibility with gelatin loadings higher that 8 wt.% is
mainly attributed to the shrinkage of the treated paper.
According to Seth [8], paper extensibility generally depends on
the stretch-potential of fibers, the degree of bonding between
fibers and the structure of the network. This is in line with our find-
ings which indicate that gelatin improves the fiber bonding
(enhanced tensile properties) and also contributes to the paper
structure (facilitates paper shrinkage during drying). Moreover,
the load–elongation curves of gelatin-modified paper indicate a
much more plastic behavior than the reference paper (control sam-
ple with no gelatin treatment). It is likely that the applied stress is
more evenly distributed in the cellulose–gelatin network, which in
turn allows the axially deformed cellulose fibers to release their
extension potential.
Fig. 4b illustrates the stress–strain curves of gelatin-treated
paper after glutaraldehyde cross-linking (see also Table 2). The
elongation at break of the treated papers increased from 11.4% to
14.1% after crosslinking with 0.5% glutaraldehyde (w/w based on
cellulose). Likewise, the tensile index increased from 70.3 to
77 N m/g. At 1% glutaraldehyde addition level the strain to failure
reached maximum value of 15%, whereas the tensile strength
remained almost unchanged when compared to that of 0.5% glutar-
aldehyde addition. However, further addition of the cross-linker
(1.5%) did not improve the mechanical properties of the treated
papers. It should be also noted that the tensile stiffness of gela-
tin-treated paper increased approximately 30% reaching a maxi-
mum value of 3.4 kN m/g upon 8 wt.% gelatin addition (Table 2).
For the comparison, the tensile stiffness index of the untreated ref-
erence sample was found to be 2.6 kN m/g. Surprisingly, the stiff-
ness properties remained unchanged upon cross-linker
application.
Glutaraldehyde is the most commonly used cross-linker with
proteins and polymers containing amine functional groups. It is
widely accepted that glutaraldehyde reacts with the residues of
amino acids, particularly with the unprotonated
e
-NH
2
functional
groups of lysine and hydroxylysine and the amino groups of the
N-terminal amino acids. Therefore, glutaraldehyde forms bonds
similar to those of Schiff bases [40]. Typically, cross-linking of gel-
atin films enhances their rigidity and tensile strength but it is also
produces a dramatic reduction in the extensibility, probably due to
the depression of molecular mobility [49]. Interestingly, in the case
of the systems discussed here the introduction of glutaraldehyde
improved the elongation. This observation can be rationalized by
the enhanced intermolecular bonding of fibers carrying adsorbed
gelatin. Therefore, the stretch capacity of the individual cellulose
fiber becomes more pronounced. In fact, it was found that the
extensibility correlated strongly with the strength of the material
(Table 2).
3.3. Formability
The formability of gelatin-treated paper was examined as a
function of the forming die temperature. The formability strain
and strength of gelatin-treated paper at two different humidity
levels (50% and 75%) are presented in Fig. 5.
With the increased processing temperatures the elongation of
the gelatin-treated papers was generally increased until reaching
a maximum (Fig. 5a and c). It is likely that the elevated tempera-
ture reduces intermolecular forces within the structure, which
increases the mobility of polymeric chains and improves their flex-
ibility [50]. However, the strength of the samples was reduced
throughout the investigated temperature range (Fig. 5b and d). It
was also observed that further increase in temperature decreases
the formability strain which may be due to the extensive softening
of wood polymers leading to the immature initiation of a fracture.
The maximum increase in elongation with temperature was
approximately 2.5% regardless of the water content of the gela-
tin-treated paper (with and without cross-linker, Fig. 5a and c). It
should be noted that the elongation also increased in the case of
the reference paper samples but to a lesser extent (increase of
1.8% and 0.85%). Therefore, it can be postulated that the disruption
of hydrogen bonding between cellulosic fibers, facilitated by
entrapped water, is more pronounced in the reference paper sam-
ples compared to the gelatin-treated paper. The humidity level,
which correlates with the water content in the sample, was found
to have a minor effect in the elongation of the samples, i.e., the gel-
atin-treated paper with higher water content only exhibited
approximately 1% larger strain compared to the samples with
lower moisture levels. This finding may be attributed to the plasti-
cizing effect of water. However, it should be noted that the maxi-
mum elongation values for gelatin-treated papers at different
moisture content were found at different processing temperatures
(Fig. 5a and c). In general, higher water content shifted the maxi-
mum elongation values to lower temperatures. This observation
is in a good agreement with the data published by Yakimets
et al. for gelatin films. They reported that by increasing the water
content from 7% to 11% the glass transition temperature of the gel-
atin films decreased from 95 to 75 °C[51].
3.4. Morphology (SEM) of gelatin-treated fibers
The morphology and microstructure of the gelatin-treated
papers was investigated by SEM (Fig. 6). As expected, the mechan-
ical refining resulted in dislocations, cracks and microcompres-
sions throughout the fiber mat (Fig. 6a) [7]. However, despite the
protein’s ability to form continuous matrices, the surface morphol-
ogy of the gelatin-treated papers remained intact upon gelatin
application (Fig. 6b and c).
Fig. 6e and f illustrates cross-section SEM micrographs of the
gelatin-treated papers. It is apparent that despite its surface appli-
cation, gelatin does not accumulate on the surface of the paper;
Fig. 7. FTIR spectra of cross-linked (bottom profile) and non-cross-linked (middle
profile) gelatin-treated papers. FTIR spectrum of cellulose is shown as a reference
(upper profile).
182 A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184

instead, it partially penetrates throughout the fiber network. Gela-
tin fills the voids between cellulosic fibers (Fig. 6e) and improves
the bonding within the system. Moreover, this leads to a more even
distribution of the applied stress, which favors the mechanical
properties of the paper. However, when glutaraldehyde crosslink-
ing is performed, gelatin penetration inside the fiber web is slightly
deteriorated (Fig. 6f) possibly due to a fast rate of cross-linking that
locks the protein in place. Thus, the surface effect is more pro-
nounced in this case.
3.5. Bulk chemical characterization (FTIR) of gelatin-treated papers
The results pointed out thus far indicate that gelatin can be
adsorbed onto cellulose fibers and the gelatin-modified papers
have improved mechanical properties. Moreover, the mechanical
properties of gelatin-treated paper can be further improved by
introducing a cross-linking agent (glutaraldehyde). This indicates
successful crosslinking reaction between glutaraldehyde and the
gelatin-treated fibers. For example, Bigi et al. reported that the glu-
taraldehyde assisted crosslinking of the gelatin films increased
their stiffness (up to 25-fold increase in the initial Young’s modu-
lus) [52]. It is likely that the crosslinking proceeds via Schiff base
(imine) formation between glutaraldehyde and the unprotonated
e
-amino groups of gelatin-treated fibers [53]. Moreover, it is possi-
ble that in gelatin-treated paper these linkages (imines) act as
energy dissipation centers which may contribute to the total
extensibility of the system. FTIR was used to investigate the nature
of the chemical bonds within the gelatin-treated paper. FTIR spec-
tra of reference cellulose paper and gelatin-treated paper with and
without cross-linker are presented in Fig. 7. As expected, the spec-
tra appeared rather similar because of the low gelatin and glutar-
aldehyde loading. However, a signal at 1540 cm
1
is only found
for gelatin-containing samples and is attributed to amide II bend-
ing vibrations of the protein molecules. Moreover, the increased
intensity of the band at about 1640–1650 cm
1
(when compared
to the reference spectrum) indicates the presence of amide I bend-
ing vibrations of gelatin. However, it should be noted that the
interpretation of this spectral region is very challenging because
of the overlapping peak from the absorbed water which is present
in all the samples.
A more pronounced peak upon spectra magnification was
observed at 1640 cm
1
, which could possibly be attributed to the
C@N stretching vibration of the imine group of Schiff bases, a struc-
ture, which should only occur after successful crosslinking of amine
groups of gelatin and glutaraldehyde (Fig. 7, bottom spectra).
Moreover, upon magnification it is possible to observe a small peak
at 916 cm
1
that is attributed to the bending vibrations of mono-
substituted alkenes which could be formed by glutaraldehyde
cross-linking of gelatin molecules [54]. FTIR demonstrated the pres-
ence (retention) of gelatin in the treated papers, which supports the
mechanical properties findings discussed before. Moreover, there is
an indication of imine formation between glutaraldehyde and gela-
tin-treated cellulose fibers, which may explain the improved
mechanical performance of the cross-linked samples.
3.6. Surface chemical analysis (XPS) of gelatin-treated paper
Table 3 summarizes the XPS results of untreated and gelatin-
treated papers with and without crosslinking. In addition, the sur-
face of glutaraldehyde-treated paper was investigated in order to
better access the chemical changes in the surface of cross-linked
gelatin-modified paper.
The surface elemental compositions of unmodified and glutaral-
dehyde-modified paper were similar and the details in the high
resolution carbon spectra indicated that glutaraldehyde is not con-
centrated on the surface of the paper substrate but rather diffuses
inside the fiber, facilitated by its low molecular weight and weak
affinity to cellulose (REF vs GLU 1%, Table 3). The amount of CAC
bonds significantly increased (from 28% to 39%) as a result of glu-
taraldehyde addition. This observation may be attributed to the
cross-linking reaction, which leads to the formation of new CAC
bonds. Moreover, the amount of nitrogen which serves as a protein
marker is higher for cross-linked paper than for the non-cross-
linked sample (16% and 11%, respectively). This indicates that
glutaraldehyde promotes gelatin retention inside the fiber web
after wet pressing.
3.7. Stamping of the gelatin-treated paper
Overall, the results presented thus far indicate that gelatin can
be used to improve the strength and extensibility of paper.
Table 3
Surface elemental composition of unmodified, gelatin-modified and cross-linked gelatin-modified paper. GLU 1% stands for 1 wt.% glutaraldehyde treated paper, GEL 4% is sample
sprayed with 4 wt.% gelatin solution and GEL 4% + GLU 1% is 4 wt.% gelatin containing sample treated with 1 wt.% glutaraldehyde solution.
Sample Atomic concentrations, % High resolution C survey C 1s spectra
C1s N1s O1s CACCAOC@O COO
Reference 60.4 ± 0.2 39.6 ± 0.2 4.4 ± 0.1 75.1 ± 0.2 18.8 ± 0.1 1.7 ± 0.1
GLU 1% 60.4 ± 0.2 39.6 ± 0.2 5.1 ± 0.1 73.2 ± 0.2 19.2 ± 0.2 2.5 ± 0.1
GEL 4% 65.0 ± 0.2 10.8 ± 0.1 24.1 ± 0.2 28.0 ± 0.2 47.4 ± 0.2 22.4 ± 0.1 2.2 ± 0.1
GEL 4% + GLU 1% 66.7 ± 0.2 15.7 ± 0.2 17.3 ± 0.1 38.7 ± 0.2 34.1 ± 0.1 24.3 ± 0.1 2.9 ± 0.1
Fig. 8. Stamped unmodified paper (a) and gelatin-treated paper (4 wt.% gelatin) (b). The dimensions of the formed shapes were approximately 110 mm (length) 70 mm
(width) 35 mm (depth).
A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184 183

Therefore, gelatin treatment can be used to produce advanced 3D-
shapes, for example, by directly stamping paper, without need of
pre-creasing. As such, paper samples were thermoformed by the
stamping method. The resultant thermoformed material, in the
shape of a trough, indicated a superior performance of the system
after gelatin treatment (Fig. 8). In fact, with no gelatin addition it is
not possible to directly process paper by this method. The process-
ability of gelatin-treated fibers is expected to also facilitate other
types of forming processes.
4. Conclusions
We demonstrate the potential of gelatin treatment in 3D-form-
ing processes for packaging applications. First, QCM-D and SPR
studies with cellulose films indicated a maximum adsorption of
gelatin at the isoelectric pH of 5.8, which was further demon-
strated by using AFM and XPS. Conditions for maximum gelatin
adsorption were employed by spraying protein solution onto paper
sheets (up to 8 wt.% gelatin content), which was noted to improve
their extensibility and strength properties. The plastic behavior
was further improved at higher gelatin loadings (over 8 wt.%) but
with a reduced strength at failure. Glutaraldehyde-assisted cross-
linking of gelatin-treated fibers further enhanced the mechanical
performance of the material. For instance, by adding 1 wt.% of glu-
taraldehyde to paper loaded with 4 wt.% gelatin increased the
elongation from 11% to 15.0% and the tensile Index from 70
to 77 N m/g.
Acknowledgments
This work was carried out under the Academy of Finland’s Cen-
tres of Excellence Programme (2014–2019) and it was financially
supported by the Finnish Bioeconomy Cluster (FiBiC LTD). Dr.
Joseph Campbell (Aalto University) is acknowledged for the
assistance in XPS measurements. Stora Enso is acknowledged for
providing the SEM images.
References
[1] J. Rhim, Food Sci. Biotechnol. 19 (1) (2010) 243–247.
[2] S. Farrisa, K. Schaich, Trends Food Sci. Technol. 20 (2009) 316–332.
[3] A. Vishtal, E. Retulainen, Bioresources 7 (3) (2012) 4424–4450.
[4] P.M. Visakh, S. Thomas, K. Oksman, P. Mathew Aji, Bioresources 7 (2) (2012)
2156–2168.
[5] M.J. John, B. Francis, K.T. Varughese, S. Thomas, Composites: Part A 39 (2008)
352–363.
[6] X. Li, L. He, H. Zhou, W. Li, W. Zha, Carbohydr. Polym. 87 (2012) 2000–2004.
[7] X. Zeng, A. Vishtal, E. Retulainen, E. Sivonen, S. Fu, Bioresources 8 (1) (2013)
472–486.
[8] R.S. Seth, Pulp Pap. Mag. Can. 106 (2) (2005) T31.
[9] S. Vuoti, E. Laatikainen, H. Heikkinen, L. Johansson, E. Saharinen, E. Retulainen,
Carbohydr. Polym. 96 (2013) 549–559.
[10] T. Nishino, I. Matsuda, K. Hirao, Macromolecules 37 (2004) 7683–7687.
[11] T. Nishino, N. Arimoto, Biomacromolecules 8 (2007) 2712–2716.
[12] M.E. Akim, G.M. Telysheva, Bumazhnaya Promyshlennost 3 (1991) 14–16.
[13] I. Levy, A. Nussinovitch, E. Shpigel, O. Shoseyov, Cellulose 9 (2002) 91–98.
[14] A. Vishtal, E. Retulainen, Nord. Pulp Pap. Res. J. 29 (3) (2014) 434–443.
[15] A. Nakayama, A. Kakugo, J. Gong, Adv. Funct. Mater. 14 (2004) 1124–1128.
[16] J. George, Carbohydr. Polym. 87 (2012) 2031–2037.
[17] R. Dash, M. Foston, A.J. Ragauskas, Carbohydr. Polym. 91 (2013) 638–645.
[18] S. Farris, J. Song, Q. Huang, J. Agric. Food Chem. 58 (2) (2010) 998–1003.
[19] A. Rokhade, Carbohydr. Polym. 65 (2006) 243–252.
[20] N. Devi, T. Maji, AAPS PharmSciTech 10 (4) (2009) 1412–1419.
[21] B. Balakrishnan, N. Joshi, R. Banerjee, J. Mater. Chem. B 1 (2013) 5564–5577.
[22] J. Ratanavaraporn, S. Damrongsakkul, N. Sanchavanakit, T. Banaprasert, S.
Kanokpanont, J. Met., Mater. Miner. 16 (2006) 31–36.
[23] Q. Xing, F. Zhao, S. Chen, J. McNamara, M. DeCoster, Acta Biomater. 6 (2010)
2132–2139.
[24] S. Rivero, M. García, Innovative Food Sci. Emerging Technol. 11 (2010) 369–
375.
[25] H. Liang, W. Chang, H. Liang, M. Lee, H. Sung, J. Biomed. Mater. Res. A 30 (1996)
361–367.
[26] R.A. Carvalho, C.R.F. Grosso, Braz. J. Chem. Eng. 23 (2006) 45–53.
[27] M. Bertoldo, S. Bronco, T. Gragnoli, F. Ciardelli, Macromol. Biosci. 7 (2007) 328–
338.
[28] J.D. Kosmala, D.B. Henthorn, L. Brannon-Peppas, Biomaterials 21 (2000) 2019–
2023.
[29] H. Liang, W. Chang, H. Liang, M. Lee, H. Sung, J. Appl. Polym. Sci. 91 (2004)
4017–4026.
[30] R.A. Carvalho, C.R.F. Grosso, Food Hydrocolloids 18 (2004) 717–726.
[31] E. Kontturi, P.C. Thuene, J.W. Niemantsverdriet, Infrared Spectrosc., At. Force
Microsc., – Langmuir 19 (14) (2003) 5735–5741.
[32] T. Tammelin, T. Saarinen, M. Österberg, J. Laine, Cellulose 13 (5) (2006) 519–
535.
[33] M. Schaub, G. Wenz, G. Wegner, A. Stein, D. Klemm, Adv. Mater. 5 (1993) 919–
922.
[34] C. Aulin, S. Ahola, P. Josefsson, T. Nishino, Y. Hirose, M. Osterberg, L. Wågberg,
Langmuir 25 (2009) 7675–7685.
[35] T. Tammelin, J. Merta, L. Johansson, P. Stenius, Langmuir 20 (25) (2004)
10900–10909.
[36] M. Rodahl, F. Höök, A. Krozer, P. Brzezinski, B. Kasemo, Rev. Sci. Instrum. 66 (7)
(1995) 3924.
[37] F. Höök, M. Rodahl, P. Brzezinski, B. Kasemo, Langmuir 14 (4) (1998) 729–734.
[38] M.V. Voinova, M. Rodahl, M. Jonson, B. Kasemo, Phys. Scr. 59 (5) (1999) 391–
396.
[39] L.S. Jung, C.T. Campbell, T.M. Chinowsky, M.N. Mar, S.S. Yee, Langmuir 14 (19)
(1998) 5636–5648.
[40] R. Schrieber, H. Gareis, Gelatine Handbook: Theory and Industrial Practice,
Wiley-VCH, Weinheim, Germany, 2007.
[41] K. Salminen, J. Kataja-aho, H. Kindqvist, E. Retulainen, T. Rantanen, A.
Sundberg, The effects of certain polymers on tensile strength and tension
relaxation of wet web, in: PaperCon. TAPPI PaperCon Conference, Covington,
KY USA, May 1–4, 2011, pp. 825–832.
[42] J. Laine, T. Lindström, G. Nordmark, Nord. Pulp Pap. Res. J. 15 (5) (2000) 520–
526.
[43] V. Kunnari, P. Jetsu, E. Retulainen, Formable paper for new packaging
applications, in: Proceedings of: 23rd Symp. International Association of
Packaging Research Institutes (IAPRI), Windsor, UK, 2007.
[44] L. Johansson, J.M. Campbell, Surf. Interface Anal. 36 (2004) 1018–1022.
[45] T.J. Su, J.R. Lu, R.K. Thomas, Z.F. Cui, J. Penfold, J. Phys. Chem. B 102 (1998)
8100–8108.
[46] E. Halder, D.K. Chattoraj, K.P. Das, Biopolymers 77 (5) (2005) 286–295.
[47] A. Kamyshny, O. Toledano, S. Magdassi, Colloids and Surf., B 13 (4) (1999) 187–
194.
[48] A. Vishtal, M. Hauptmann, R. Zelm, J. Majschak, E. Retulainen, Packag. Technol.
Sci. 27 (2013) 677–691.
[49] J. Martucci, J. Mater. Sci. 47 (2012) 3282–3292.
[50] A. Gregorova, J. Lahti, R. Schennach, F. Stelzer, Thermochim. Acta 570 (2013)
33–40.
[51] I. Yakimets, S. Paes, Biomacromolecules 8 (2007) 1710–1722.
[52] A. Bigi, G. Cojazzi, S. Panzavolta, K. Rubini, N. Roveri, Biomaterials 22 (2001)
763–768.
[53] L.H.H. Olde Damink, P.J. Dijkstra, M.J.A. van Luyn, P.B. van Wachem, P.
Nieuwenhuis, J. Feijen, J. Mater. Sci.: Mater. Med. 6 (1995) 460–472.
[54] I. Migneault, C. Dartiguenave, M.J. Bertrand, K.C. Waldron, Biotechniques 37
(5) (2004) 790–802.
184 A. Khakalo et al. / Reactive & Functional Polymers 85 (2014) 175–184