The effect of oxyalkylation and application of polymer dispersions on the thermoformability and extensibility of paper

Carbohydrate Polymers 186 (2018) 411–419

Contents lists available at ScienceDirect

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

The effect of oxyalkylation and application of polymer dispersions on the
thermoformability and extensibility of paper

T

⁎

Jarmo Koukoa, , Harri Setäläb, Atsushi Tanakab, Alexey Khakalob, Jarmo Ropponenb,
Elias Retulainena
a
b

VTT Technical Research Centre of Finland Ltd, P.O. Box 1603, FI-40101 Jyväskylä, Finland
VTT Technical Research Centre of Finland Ltd, P.O. Box 1000, FI-02044 VTT, Finland

A R T I C L E I N F O

A B S T R A C T

Keywords:
Biopolymer spraying
Consumer packages
Paper extensibility
Starch acetate

Strength of paper
Thermoformable web

Wood fiber-based packaging materials, as renewable materials, have growing market potential due to their
sustainability. A new breakthrough in cellulose-based packaging requires some improvement in the mechanical
properties of paper. Bleached softwood kraft pulp was mechanically treated, in two stages, using high- and lowconsistency refining, sequentially. Chemical treatment of pulp using the oxyalkylation method was applied to
modify a portion of fiber material, especially the fiber surface, and its compatibility with polymer dispersions
including one carbohydrate polymer. The results showed that the compatibility of the cellulosic fibers with some
polymers could be improved with oxyalkylation. By adjusting mechanical and chemical treatments, and the
thermoforming conditions, the formability of paper was improved, but simultaneously the strength and stiffness
decreased. The results suggest that the formability of the paper is not a direct function of the extensibility of the
applied polymer, but also depends on the fiber network structure and surface energy.

1. Introduction

cellulosic fibers strong and stiff. However, mechanical treatment at high
consistency possibly combined with a low consistency refining phase
has been shown to improve the elongation potential of paper (Khakalo,
Vishtal, Retulainen, Filpponen, & Rojas, 2017; Sjöberg & Höglund,
2005; Zeng, Vishtal, Retulainen, Sivonen, & Fu, 2013). In this study, the
influence of combined mechanical high consistency – low consistency
treatment, chemical (oxyalkylation) treatment, and application of
thermoplastic and carbohydrate polymer dispersions on the formability
of bleached kraft softwood pulp, was investigated. The objective of the
treatments was to modify the bonding ability of the fiber surface and
change the fiber shape and morphology in order to improve the elongation and bonding ability of the fibers.
Elongation of some thermoplastic polymers can reach 400–800%
and therefore, it is reasonable to expect that the addition of such
polymers to the fiber network will improve the formability of the paper
(Waterhouse, 1976). Bio-based thermoplastic polymers are generally

not hazardous to health and are also bio-degradable, which makes them
suitable for use in food packages. Challenges of polymer applications to
the pulp suspension are low retention in the fiber network and insufficient adhesion to the fibers. On the other hand, in cases where the
polymer is applied on a formed fiber network, the retention is a less
severe problem, but difficulties arise in the limited penetration of the
polymer into the fiber network, and possibly in the limited adhesion

Paper-based packaging materials, as renewable materials, have a
growing market potential due to their sustainability. However, the
development of new packaging concepts requires improvement in the
mechanical properties of paper. High extensibility is one of these
properties. Highly extensible papers would have the potential to replace
certain kinds of plastics used in packaging.
Formability can be defined as the ability of a material to deform
without breaking. However, formability is not a specific mechanical
property, but can be regarded as a generic term for explaining how well
paper deforms during a particular forming process (Vishtal, 2015). In
this study, formability was mainly estimated on the basis of a 2D experimental test method that simulated the process conditions in a fixed
blank thermoforming process (Vishtal & Retulainen, 2014). In the fixed
blank process, the formability is determined by the extensibility and
tensile strength of paper (Östlund, Borodulina, & Östlund, 2011;
Vishtal, Hauptmann, Zelm, Majschak, & Retulainen, 2013). As yet, the
fixed blank forming process has not been widely applied in industry for
paperboard (Ford, Trott, Simms, & Hartmann, 2014; Vishtal 2015).
Pulp fibers constitute the load-bearing components of paper. Kraft
pulp fibers primarily consist of cellulose and hemicellulose. Cellulose is
crystalline, strong and stiff material with low extensibility making

⁎

Corresponding author.
E-mail address: jarmo.kouko@vtt.fi (J. Kouko).

/>Received 12 October 2017; Received in revised form 12 January 2018; Accepted 20 January 2018
0144-8617/ © 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( />

Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

Fig. 1. 13C CP/MAS NMR spectra of the H-substituted (A) and L-substituted (B) pulps.

and compatibility of the polymer with the fiber. Therefore, fibers were
also chemically modified by using etherification with butyl glycidyl
ether, which produces a substituent with a longer side-chain, 3-butoxy2-hydroxypropyl. This kind of substituent has been observed to improve
the thermo-mechanical behavior of cellulose fibers (Zhang, Li, Li,
Gibril, & Yu, 2014) and was also expected to improve the compatibility
of the fiber surfaces with the applied polymers.
Several thermoplastic polymers were applied onto, and into, the
prepared undried fiber networks and then the influence of the application on the formability of the paper was investigated.
Temperature is known to increase the elongation and decrease the
tensile strength and tensile stiffness of paper (Back & Andersson, 1992;
Kouko, Retulainen, & Kekko, 2014; Salmén & Back, 1980; Salmén,
1993). In this study, the influence of temperature on the formability of
paper was measured using an experimental method that simulated the
process conditions in thermoforming. The influence of the moisture
content of the paper was controlled, but not varied, though moisture is
known to act as a softener, increasing the elongation and decreasing
tensile strength and tensile stiffness (Andersson & Berkyto, 1951).
The objective was to produce a thermoformable network composed

of cellulosic fibers and bio-based thermoplastic polymers with high
formability. This study presents the results of laboratory scale methods
to improve the extensibility of paper. The extensibility of paper was
tested with a tailor-made laboratory scale device that simulated process
conditions in thermoforming.

2.2. Oxyalkylation of the pulp
Chemical modification, an oxyalkylation treatment, was applied to
modify the fiber material, especially the fiber surface, to introduce
substituents with longer side chains. The treatment was carried out onto
two levels of substitution in the alkali reaction conditions, using
methods similar to those published by Tian, Ju, Zhang, Duan, and Dong
(2015). This was done in order to modify the properties of the fibers
and their compatibility with polymer dispersions.
800 mL of deionized water and 850 mL of 90% aqueous tBuOH were
added to a 5 L glass reactor. 500 g of pulp containing 182 g of cellulose
(1.12 mol of anhydroglucose units, AGU) were added to the reactor,
and then 160 mL of 10 M NaOH was added. The reaction mixture was
stirred overnight at 45 °C. 400 mL (2.80 mol) or 926 mL (6.47 mol) of
BGE was added for the preparation of the higher (H) or lower (L)
substituted fiber samples. The reaction mixture was again stirred
overnight at 45 °C. The reaction mixture was cooled down to room
temperature and neutralized with 37% HCl. The fibers were filtrated,
then washed with 2 L of 95% ethanol, 2 L of 50% aqueous ethanol, 2 L
of 20% aqueous ethanol, and finally two times with 2 L of deionized
water.
A small portion of the products was freeze-dried for analytical
purposes. The degree of substitution (DS) of 3-butoxy-2-hydroxypropyl
group was determined using 13C CP/MAS acquired with a 600 MHz
Agilent NMR spectroscope. The NMR spectra of H-substituted (DS 0.12)

and L-substituted (DS 0.05) fiber samples are presented in Fig. 1. As a
reference (R), the high and low consistency refined BSKP was also
studied without the chemical treatments. Investigated pulp samples are
presented in Table S1.

2. Experimental
2.1. Raw materials for the pulp preparation

2.3. Paper sheet preparation and spraying of the polymers
Bleached softwood kraft pulp (BSKP) from a Finnish mill was used
as the fiber raw material. The BSKP was mechanically treated, in two
sequential stages, using a high-consistency mechanical treatment (with
a wing defibrator) and a low-consistency refining method (with a Valley
beater), as it is known to improve the extensibility of paper (Khakalo
et al., 2017; Sjöberg & Höglund, 2005; Zeng et al., 2013).
Butyl glycidyl ether (95%, BGE), tert-butanol (99%, tBuOH), 37%
hydrochloric acid solution, and sodium hydroxide solution (10 M,
NaOH) were purchased from Sigma-Aldrich. All chemicals were used as
received.

The treated pulps were made into 60 g/m2 laboratory sheets using
EN ISO 5269-3. Laboratory sheets were wet pressed at 350 kPa pressure
(the EN ISO 5269-3).
The selected polymer dispersions (see Table 1) were sprayed onto
the top side of the wet sheets using a commercial, high-volume, lowpressure (HVLP) gravity feed air spray gun. For spraying, a wet sheet
was placed on a rigid plastic plate that was on a gravimetric scale. The
targeted amount of dispersion for a sheet was 20%, and it was adjusted
by controlling the wet weight of the sheet after spraying based on the
known consistency of the dispersion. The targeted amount of a polymer
412

Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

(PPC) at a 25wt% dispersion. Miyoshi Oil & Fat co. Ltd. supplied the
polylactic acid Landy-3000 emulsion (PLA) at 40wt% dispersion (the
average particle size was 1 μm). Sigma Aldrich’s gelatin (GL) from cold
water fish skin (G7041) was diluted with water to a 2wt% dispersion.
Oriola Oy’s citric acid (CA) was obtained as a powder and diluted with
water to a 40wt% solution. The addition of gelatin has been found to
improve paper formability (Khakalo et al., 2014), which was the motivation for using it with PLA (GL-PLA). On the other hand, citric acid is
generally known to act as a softener for PLA and therefore it (PLA-CA)
was expected to improve the extensibility of the polymer-impregnated
fiber network. Emerald Performance Materialsđ supplied the nitrile
latex Nychemđ 1561 ì 604 (NLAT) as a 33wt% emulsion. Pure water
was used as a reference for the polymers. All of the sprayed polymer
dispersions and their application consistencies are presented in Table 1.

Table 1
The sprayed polymers.
Polymer Abbreviation

Polymer Name

Consistency, %

WREF
PU-DL*

PU-EPO
PPC
PLA
GL-PLA

Water (reference)
Polyurethane (Impranil® DL 519)
Polyurethane (Epotal® P 100)
Polypropylene carbonate
PLA (LANDY PL-3000)
Gelatin (2% of dry wt.)+
PLA (18% of dry wt.)
PLA + citric acid (4:1 mixture)
Starch acetate
Nitrile latex

–
40
39
25
40
2 and 40

PLA-CA
ST-AC
NLAT

40
7.8
33

The dry weights of the sprayed polymer mixtures were equal to 20% of dry fiber amount,
except gelatin was first sprayed equal to 2% and after that PLA equal to 18% of the dry
fiber amount.
* Polymer sprayed onto the board samples (in the Supplementary material).

2.6. Dynamic mechanical analysis (DMA) of the polymers and papers
Polymer film samples were made from the dispersions studied in
order to determine their softening behavior. Dynamic mechanical
analysis (DMA) was performed using a DMA/SDTA 861 E Dynamic
Mechanical Analyzer (Mettler-Toledo Inc.) at VTT. The storage modulus
and phase lag values were measured in shear mode as a function of
temperature. The temperature range in this investigation was from −40
to 120 °C. The maximum of the loss modulus at the studied temperature
range can be roughly regarded as a softening temperature of a tested
material. Paper samples were tested in tensile mode.

was then sprayed, as evenly as possible, onto the sheet. Reference
sheets were sprayed with an equivalent amount of pure water (WREF).
The sprayed samples were set on a vacuum table covered with a
fabric wire; a vacuum level of approximately 5–10 kPa was applied
under the sample. The aim of the vacuum treatment was to impregnate
the polymers into the wet fiber network and obtain the potential advantages of not only a paper-polymer composite, but also of a fiberpolymer composite structure.
The sheets were dried, without restraint, in between two polymer
wires with a gap of approximately 3 millimeters. The applied drying
method enabled free shrinkage of the sheet structure during drying in
the in-plane direction, while simultaneously preventing severe cockling
and curling of the sheets. The purpose of the unrestrained drying was to
impose a high level of extensibility in the formability tests. The obtained polymer amounts were estimated based on the basis weight of
the dry paper samples.

2.7. Mechanical tests of the papers
The basis weight of the paper (the board sample described in the
Appendices A1) samples was determined according to ISO 536:1995.
The thickness was determined according to ISO 534:1998, and the
density was determined based on the measured values of the basis
weight and thickness. The tensile strength and the strain at break were
determined with a Lloyd tensile tester, in accordance with ISO
5270:1998. Paper samples were conditioned and all testing of the
samples took place at a temperature of 23 °C and at 50% relative humidity.

2.4. Preparation of starch acetate (carbohydrate) dispersion for spraying
The acetylated potato starch acetate DS 2.8 was obtained from the
VTT Technical Research Centre of Finland (Rajamäki, Finland). The
TECOSA was prepared in a 3:2 ratio from triethyl citrate (TEC) and noctenyl succinate anhydride (OSA). TEC was purchased from Reilly
Chemicals, Hautrage, Belgium and OSA from Pentagon Chemicals,
Cumbria, USA. All chemicals were used as received.
The aqueous starch acetate (carbohydrate) dispersion (ST-AC, see
Table 1) was prepared as described by Mikkonen, Peltonen, Heikkilä,
and Hamara (2000). 100.0 g of an acetylated potato starch (DS 2.8),
Mowiol (12.0 g), TECOSA (80.0 g) and 12 g of water were melted at
100 °C for 4 h, in a flask equipped with a mechanical stirrer. Then, the
temperature was adjusted to 95 °C for 1.5 h. Stirring was continued for
the next 1.5 h, while 40.0 g of water at 80 °C was slowly added. The
dispersion was allowed to cool afterwards to room temperature, while
under continuous stirring. The dispersion was allowed to stand overnight without stirring, after which the final dispersion was obtained at
61% solids content. In order to enable lower viscosity for the spraying,
the starch acetate (ST-AC) was diluted with water to a dispersion at 7.8
wt% consistency.

2.8. 2D formability test
The formability strain and force of the paper samples were measured
using a 2D-formability tester at VTT. The measurement procedure set-up
was described by Vishtal and Retulainen (2014) and illustrated in Fig. S1.
In this investigation, the velocity of the forming press was 1 mm/s and
the width of the paper sample was 20 mm. The paper samples were set in
the tester so that the sprayed surface was not in contact with the heated
press. The temperature of the paper samples with 60 g/m2 was measured
with the lower infrared thermometer. Paper sample temperatures during
the formability tests were typically only 2–5 °C lower than the presented
set temperature, which means there was only a minor temperature gradient through the paper thickness.
2.9. Contact angle and surface energy tests
Surface-free energy of papers made from modified pulp was determined from contact angle (CA) measurements with water, formamide, diiodomethane, and ethylene glycol. The CA of sessile drops
was determined using a KSV CAM200 optical contact angle goniometer
(KSV Instruments). Surface-free energy parameters for the probing liquids used for the CA measurements are available in the literature (van
Oss, Good, & Chaundhury, 1988). The CA values (average of three
measurements) were then used to calculate the dispersive and polar
contributions to the surface energy of the samples, according to the
acid–base theory (Good, Girifalco, & Kraus, 1958; Good & Girifalco,
1960).

2.5. Description of the other sprayed polymers
Two commercial polyurethane water dispersions were used in this
study. The Impranil® DL 519 polyurethane dispersion (PU-DL) was
supplied by Bayer AG (now Covestro) as a 40wt% dispersion in water
(the average particle size was 110 nm). The Epotal® P 100 Eco (PU-EPO)
was kindly supplied by BASF SE as a 40wt% dispersion in water the
average particle size distribution was below 100 nm. Empower
Materials supplied the polypropylene carbonate QPAC® 40 emulsion
413

Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

Table 2
The Softening of the polymers by DMA measurement.
Polymer
abbreviation

PU-DL
PU-EPO
PPC
PLA
ST-AC
NLAT

Polymer name

Estimated softening temperature and value
of the correspondent phase lag (tan δ), °C

®

Polyurethane (Impranil DL 519)
Polyurethane (Epotal® P 100)
Polypropylene carbonate
PLA (LANDY PL-3000)
Starch acetate

Nitrile latex

Storage modulus, MPa

°C

δ, °

23 °C

80 °C

78
60
39
74
45
38

21.8
27.5
58.5
19.3
39.7
45.0

53
86
147
130

∼40
265

4.65
0.13
0.16
1.94
0.13
0.89

Table S2. The fiber properties were measured using an L&W STFI
FiberMaster. The chemical modification treatment slightly decreased
the fiber length from 1.80 mm to 1.71 mm, increased the fiber curl and
stiffness (see shape factor from 85.4 to 83 and bendability from 6.65 to
5.4, respectively) and increased the amount of fines fraction from 8.5%
to 14%. Both the shape factor and the amount of fines have the potential to increase fiber network elongation.

2.10. Nip peeling test
The delamination resistance of double layer handsheets glued together with selected polymers was measured using a nip peeling test
(Tanaka, Kettunen, Niskanen, & Keitanniemi, 2000). Two synchronized
rolls, 25 mm in radius, were attached to an ordinary tensile tester,
which rotates the upper roll. Adhesive tape was used at the beginning of
the specimen to adhere the first 10–20 mm of the specimen to the rolls.
After the beginning, peeling proceeded without the tape.
The nip peeling test was performed for the three pulps (R, L and H),
but only with the PU-DL and PU-EPO polymers. PU was sprayed on a
dry sheet and, after spraying, another identical sheet was gently set on
the sprayed sheet and set under a metal plate. The amount of sprayed
PU was equal to a 40% wt. of the dry weight of both of the laboratory
sheets. The samples were dried using a drum dryer at 70 °C.

3.2. Polymer contents
The polymer contents in the laboratory sheets are presented in Fig.
S2. In the L-substituted pulp samples, the polymer amount was systematically 6–8wt% higher than in the unmodified reference (R) or the
H-substituted samples. Therefore, the WREF (water sprayed reference)
samples had a key role, when the different samples were compared. The
estimated polymer amounts were based on the mechanical properties of
the paper samples that are presented in Table S3.
Polymer amounts were very similar for the PU (polyurethane) and
NLAT (nitrile latex) sheets. Additionally, the polymer content was similar for all of the PLA sheets (PLA, GL-PLA and PLA-CA). The sheets
with PPC (polypropylene carbonate) and ST-AC (starch acetate) had
polymer amounts that were significantly lower in comparison to the
other sheets.

2.11. SEM imaging
The SEM imaging was carried out with a Merlin FE-SEM (Carl Zeiss
NTS GmbH, Germany) with gold sputter coating at 20 mA for 30 s. First,
the sample was attached onto an aluminium specimen stub with a
double-sided carbon adhesive tape. The imaging was performed using
1.5 keV electron energy using a secondary electron detector. The image
pixel resolution was 2048 × 1536.
The SEM imaging was performed for the three pulps (R, L and H),
but only with the water (reference) and starch acetate spraying, excluding the L-substituted sample with water spraying.

3.3. Dynamic mechanical analysis (DMA) of the polymers
The films made of the applied polymers were studied using DMA in
shear mode. Increasing temperature softened the polymers and caused
significant changes in the storage modulus, loss modulus and phase lag
values. The estimated softening temperature with the corresponding
phase lag and storage modulus values at 23 °C and 80 °C are presented

in Table 2. The storage modulus at 80 °C was expected to have a connection with the behavior of the polymers in paper samples during the
thermoforming process.
The observed softening temperatures of the polymers were between
38 °C and 78 °C and the observed phase lag values were between 19.3°
and 61.7°. As is well-known, a perfectly elastic solid material and a
purely viscous fluid have phase lags 0° and 90°, respectively. In case of a
gel, the phase lag should be less than 45°. Polypropylene carbonate (PPC)
and PU-EPO (at 120 °C) had the highest phase lags (approximately 60°)
indicating that those polymers can reach the most fluid-like behavior of
the studied polymers. The fluidization of the polymer by temperature
may be very important for the sliding of individual fiber contacts and
overall flexibility of the fiber network during the thermoforming.
A DMA test was also performed on the pulp samples, but the oxyalkylation treatment had no effect on the log-linear decay of the storage
modulus, and there was no maximum for the phase lag.

3. Results and discussion
3.1. Chemical modification of pulp fibers
The reaction efficiency of oxyalkylation reaction was rather low
(only approximately 2%) and DSes were only 0.05 (L) and 0.12 (H).
However, the reaction efficiency and DS were similar to results reported
earlier. When Vehviläinen et al. (2015) derivatized cellulose fibers
using allyl glycidyl ether under similar heterogeneous reaction conditions and Qi, Liebert, & Heinze (2012) in homogeneous reaction conditions like in NaOH/urea, the reaction efficiencies were only 1–3% and
DSallyl were approximately 0.2. They also observed that when DSallyl
was 0.22–0.29, the cellulose derivatives were only swelling and when
DSallyl was above 0.50 they were already soluble in water. Nishimura,
Donkai, & Miyamoto (1997) have prepared similar 3-butoxy-2-hydroxypropyl cellulose derivatives and they have also observed good solubility in water when DS was sufficiently higher such as 0.4-1.0. Due to
those results, the target of DS in this study was under 0.2 to keep the
product in a fibrous form. Because both of the DSes were rather low
(< 0.15), the modified cellulose products L and H were still in a fibrous
form. However, already this rather light chemical modification and

low-level DSs can, evidently, change fiber properties and, for example,
to improve their compatibility with other polymeric materials.
The fiber properties of the chemically treated pulps are presented in

3.4. Surface energy of the pulps
The surface energy of the pulps was expected to affect the fibers and
their bonding ability and their compatibility with the polymers. The
414


Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

test results, as the contribution ratio was around 30% (see Frigon and
Mathews (1997) for contribution ratio). This means that the systematically better strain at break of L-substituted samples has been partly
caused by the higher polymer level in the sheets.
It is known that the shape factor of fibers and fines content influence
the elongation of paper. In this investigation, the shape factor and fines
content were changed with the chemical modification of the pulp (see
Table S2.), which could be studied by comparing the WREF samples.
However, strain at break of the R, L and H pulps did not seem to have
the correlation with the fines content and shape factor. The WREF of
the H-substituted pulp had the lowest elongation, which indicates that
the chemical modifications seem to contribute to the formability of the
paper at room temperature.
The tensile index and elastic modulus were clearly decreased with
increased oxyalkylation levels (presented in Figs. 3 and S3, respectively).
A simple regression analysis (independent variables were pulp type,
chemical type and chemical amount) showed that the pulp type, alone,

explained around 70% of the statistical variation of the tensile index and
elastic modulus. Probably the reason for the reduction was the reduced
strength of inter-fiber bonding. For the formability potential of paper, the
importance of the tensile index, and especially the elastic modulus, can
be regarded as secondary, compared to strain at break.
The tension-strain curves of the WREF paper samples are presented
in Fig. 4. The presented curves are repetitions that closely represent the
measured averages of the strain at break and tensile index. The tensionstrain curves of the sprayed L-substituted pulp sheets are presented in
Fig. S4. The samples containing PLA generally showed good tension
levels, compared to the average performance of the samples. Moreover,
incorporation of only 2wt% of gelatin (sample L-substituted GL-PLA)
leads to even further improvement of mechanical properties and extensibility. Beneficial action of gelatin towards paper formability improvement was described in more detail in our previous communication
(Khakalo et al., 2014).

Table 3
The surface energy of the pulps.
Pulp sample

γLW

γ+

γ−

γAB

γS

Reference
L-substituted

H-substituted

42.7
28.4
23.2

0.1
0.4
4.2

50.3
0.3
0

3.5
0.7
0.5

46.2
29.1
23.7

γLW – dispersive component (hydrophobic interactions).
γAB – polar component (hydrophilic interactions).
γS – total surface energy.

surface energies of the studied pulps gave a clear indication that the
surface character of the modified fibers had been changed into direction
of a higher hydrophobicity as presented in Table 3. The corresponding
polar and dispersive contributions to the surface energy shown in

Table 3 indicate that the major change in the fiber surfaces was associated with an almost total elimination of the polar contribution, although the dispersive counterpart was also critically reduced. These
results indicate that the oxyalkylated fibers may have better compatibility with thermoplastic polymers.
3.5. Tensile properties at standard test room conditions
Fig. 2 shows that the lower (L-substituted) oxyalkylation level
yielded, on average, approximately 12% strain at break (i.e., strain at
the maximum load), which was the best average among the studied
pulps (R, L, and H). The unmodified reference (R) and the higher (Hsubstituted) oxyalkylation samples had an average strain at break of
approximately 9%. However, the water sprayed (WREF) unmodified
reference (R) pulp sample clearly had a higher strain at break compared
to the WREF of the H-substituted sample, which indicates that a high
level of oxyalkylation is not beneficial for extensibility and formability
at room temperature.
Strain at break of the WREF unmodified reference (R) pulp and
modified L-substituted pulp was on the same level as the best polymers
of their groups. On the other hand, the starch acetate (ST-AC), which is
an agro-based carbohydrate polymer, had a poor strain at break,
compared to the other trial points, in all three cases. However, a simple
linear regression analysis (the independent variables were the pulp
type, chemical type and chemical amount) showed that the amount and
type of the sprayed chemical had a significant influence on the tensile

3.6. 2D formability strain
2D formability strain was studied at four different press temperatures (23 °C, 60 °C, 90 °C and 120 °C). Figs. 5, S5 and S6 show that the
maximum formability strain was typically obtained at 60 °C temperature, whereas 120 °C clearly seemed to be above the optimal temperature.
Fig. 2. Strain at break of the paper samples.

415


Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

Fig. 3. Tensile index of the paper samples.

The water sprayed reference samples (WREF) of the unmodified (R)
pulp and modified L-substituted (oxyalkylated) samples had a clearly
higher formability strain compared to the H-substituted (oxyalkylated)
sample, which once again indicated that high oxyalkylation level may
not be beneficial for high formability.
The influence of polymer spraying on the 2D formability strain of
the chemically unmodified reference (R) pulp was generally quite
negligible. Only PU-EP and NLAT seemed to improve the 2D formability
strain of the reference pulp. On the other hand, the three PLA trial
points, and especially the PPC and ST-AC (the only carbohydrate
polymer) trial points, had a significantly lower amount of polymers
than the PUs and NLAT.
Several polymers improved the 2D formability strain of the L-substituted (oxyalkylated) samples, compared to the water sprayed WREF
sample. Both of the PUs (PU-DL and PU-EP), all three PLA mixtures
(PLA, GL-PLA and PLA-CA) and the nitrile latex (NLAT) improved the
2D formability strain. However, the PLA sheets contained 4–6% wt. less
polymer than the PUs and NLAT sheets, and therefore, PLA and PU may
have similar potential for improving the formability of paper.

Fig. 4. Tension-strain curves of the WREF paper samples, of the investigated pulps. In the
legend, the samples are presented in the order that matches the order of magnitude at 6%
strain.

Fig. 5. Formability strain of the paper samples with L-substituted (oxyalkylated) pulp.

416


Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

with increasing temperature (23 °C, 60 °C, 90 °C and 120 °C). Maximum
formability forces were measured at the lowest 23 °C temperature,
while the 2D formability forces were 25–75% of the maximum value at
120 °C. The maximum formability strain was obtained at around 60 °C
and therefore, the 2D formability test results were compared at that
temperature in the following analysis.
The influence of the sprayed polymers was minor on the 2D formability force of the reference (R) samples. However, the average level
(150 N at 60 °C), of the reference (R) sample 2D formability force, was
higher compared to the L- and H-substituted (oxyalkylated) samples.
In the case of the L-substituted and H-substituted samples, the
sprayed polymers increased the formability force. The average 2D
formability forces were 115 N and 60 N for the L- and H-substituted
pulps at 60 °C, respectively. Both temperature and polymers had a
major influence on the 2D formability force.
The 2D formability test results showed that by choosing the treatment method, the chemical composition drying method and thermoforming conditions, the extensibility of a dried BSKP sample reached
the 16% level. The results also indicated that the formability of the
paper is not a direct function of the extensibility of the applied polymer.

Fig. 6. Peeling energy of the PU sprayed samples.

Most of the sprayed polymers improved the 2D formability strain of
H-substituted samples. However, the average 2D formability strain of
the H-substituted samples was significantly lower compared to the reference (R) and the L-substituted samples.

Figs. S7–S9 show that the 2D formability force strongly decreased

Fig. 7. SEM images of untreated (R), lower (L) and
higher (H) substituted (oxyalkylated) BSKP samples.
Images a–b present water sprayed (WREF) and c-e
present starch acetate sprayed (ST-AC) samples.

417


Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

3.7. Delamination resistance

Acknowledgements

The compatibility of the pulps with polyurethane was studied by
testing the delamination resistance of two-ply handsheets bonded together with polyurethane. The results presented in Fig. 6 show that the
peeling energy of the H-substituted (oxyalkylated) pulp was clearly
lower than that of the chemically unmodified reference (R) and the Lsubstituted (oxyalkylated) pulp. The L-substituted pulp had the highest
peeling energy. The difference between the PU grades was minor,
compared to the standard deviation of the measurements. The results,
once again, indicated that the L-substitution treatment improved the
compatibility of the PU and the fibers, i.e., adhesion, while an excessively heavy chemical treatment, was unfavorable for the formability of the paper.

This work was a part of the ACel program of the Finnish
Bioeconomy Cluster CLIC Innovation. Funding by the Finnish Funding
Agency for Technology and Innovation (TEKES) is gratefully acknowledged. Completion of the publication was supported by the ExtBioNet

project funded by the Academy of Finland. Dr. Oleg Timofeev is
thanked for producing the polymer film samples and Ms. Mirja Nygård
for conducting the DMA measurements from them.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at />References

3.8. SEM imaging
Andersson, O., & Berkyto, E. (1951). Some factors affecting the stress–strain characteristics of paper. Svensk Pappers Tidning, 54(13), 437–444.
Back, E. L., & Andersson, L. I. (1992). The effect of temperature on wet web strength
properties. Pan-Pacific Pulp and Paper Technology, B, 141–150.
Ford, N., Trott, P., Simms, C., & Hartmann, D. (2014). Case analysis of innovation in the
packaging industry using the cyclic innovation model. International Journal of
Innovation Management, 18(5), />1450033-24.
Frigon, N. L., & Mathews, D. (1997). Guide to experimental design. John Wiley & Sons, Inc.
342 p.
Good, R. J., & Girifalco, L. A. (1960). A theory for estimation of interfacial energies. III.
Estimation of surface solids from contact angle data. Journal of Physical Chemistry,
64(5), 561–565.
Good, R. J., Girifalco, L. A., & Kraus, G. (1958). A theory for estimation of interfacial
energies. II. Application to surface thermodynamics of teflon and graphite. Journal of
Physical Chemistry, 62(11), 1418–1421.
Khakalo, A., Filpponen, I., Johansson, L.-S., Vishtal, A., Lokanathan, A. R., Rojas, O. J.,
et al. (2014). Using gelatin protein to facilitate paper thermoformability. Reactive &
Functional Polymers, 85, 175–184. />2014.09.024.
Khakalo, A., Vishtal, A., Retulainen, E., Filpponen, I., & Rojas, O. (2017). Mechanicallyinduced dimensional extensibility towards tough fiber networks. Cellulose, 24(1),
191–205. />Kouko, J., Retulainen, E., & Kekko, P. (2014). Straining and relaxation properties of wet
paper during heating. Mechanics of Time-Dependent Materials, 18(4), 697–719. http://
dx.doi.org/10.1007/s11043-014-9246-4.
Mikkonen, H., Peltonen, S., Heikkilä, M. E., & Hamara, J. (2000). Process for the preparation of polymer dispersions. Patent FI, 105566.

Nishimura, H., Donkai, N., & Miyamoto, T. (1997). Preparation and properties of a new
type of comb-shaped, amphiphilic cellulose derivative. Cellulose, 4(2), 89–98. http://
dx.doi.org/10.1023/A.1018467202853.
Qi, H., Liebert, T., & Heinze, T. (2012). Homogenous synthesis of 3-allyloxy-2-hydroxypropyl-cellulose in NaOH/urea aqueous system. Cellulose, 19(3), 925–932. http://
dx.doi.org/10.1007/s10570-012-9687-3.
Salmén, L., & Back, E. L. (1980). Moisture dependent thermal softening of paper evaluated by its elastic modulus. Tappi, 63(6), 117–120.
Salmén, L. (1993). Responses of paper properties to changes in moisture content and
temperature. In C. F. Baker (Vol. Ed.), Products of papermaking: vol. 1. Leatherhead:
Pira International.
Sjöberg, J., & Höglund, H. (2005). Refining system for sack paper pulp: Part I HC refining
under pressurised conditions and subsequent LC refining. Nordic Pulp & Paper
Research Journal, 20(3), 320–328.
Tanaka, A., Kettunen, H., Niskanen, K., & Keitanniemi, K. (2000). Comparison of energy
dissipation in the out-of-plane and in-plane fracture of paper. Journal of Pulp and
Paper Science, 28(11), 385–390.
Tian, Y., Ju, B., Zhang, S., Duan, X., & Dong, D. (2015). Preparation and phase transition
behaviors of temperature-responsive 3-butoxy-2-hydroxypropyl hydroxyethyl celluloses. Journal of Biomaterials Science, Polymer Edition, 26(16), 1100–1111. http://dx.
doi.org/10.1080/09205063.2015.1077918.
Van Oss, C. J., Good, R. J., & Chaundhury, M. K. (1988). Additive and nonadditive surface
tension components and the interpretation of contact angles. Langmuir, 4(4),
884–891. />Vehviläinen, M., Kamppuri, T., Setälä, H., Grönqvist, S., Rissanen, M., Honkanen, M.,
et al. (2015). Regeneration of fibres from alkaline solution containing enzyme-treated
3-allyloxy-2-hydroxypropyl substituted cellulose. Cellulose, 22(4), 2271–2282.
/>Vishtal, A., & Retulainen, E. (2014). Improving the extensibility, wet web and dry
strength of paper by addition of agar. Nordic Pulp & Paper Research Journal, 29(3),
434–443. />Vishtal, A., Hauptmann, M., Zelm, R., Majschak, J. P., & Retulainen, E. (2013). 3D
forming of paperboard: The influence of paperboard properties on formability.
Packaging Technology and Science, 27(9), 677–691. />2056 wileyonlinelibrary.com.

SEM images of BSKP sheet surfaces are presented in Fig. 7a–e.

Fig. 7c–e showed that the starch acetate did not penetrate the cell wall
of fibers, but rather accumulated on the surface. Also, starch seems to
be more prone to spread onto the fiber surface of the reference (DS 0)
sample (Fig. 7c), whereas no spreading of starch acetate was observed
onto L- (DS 0.05) or H-substituted (DS 0.12) (oxyalkylated) samples
(Fig. 7d–e, respectively). The behavior of starch acetate (ST-AC) in
these SEM images can probably be explained by the surface energy
measurements, which showed that hydrophobicity of the sheet surface
was increased with oxyalkylation, i.e., the substitution DS level. Consistency of the starch acetate in spraying was 7.8%, whereas the amount
of water was over 90%. As a result of hydrophobicity of the BSKP sheet
surface and high water content, the starch acetate was probably forced
to form from spherical particles after drying. The starch acetate was
slightly able to spread on the untreated BSKP surface (Fig. 7c). In case
of the starch acetate, this observed behavior probably explains the inability of starch acetate to improve formability of BSKP sheets. The
compatibility of the other polymers with BSKP sheet may also have
been influenced by the high water content of the spray dilutions.

4. Conclusions
Bleached softwood kraft pulp was mechanically treated, in two
stages, using high- and low-consistency refining, sequentially. Chemical
treatment of pulp using the oxyalkylation method was applied in order
to modify the fiber material, especially the fiber surface, and its compatibility with polymer dispersions including one carbohydrate
polymer.
Oxyalkylation of the BSKP to the lower substitution (DS 0.05) level
increases hydrofobication and also improves the extensibility of the
fiber network, but not the tensile strength. The chemical oxyalkylation
modification increases surface energy of the fiber network, which can
improve the compatibility of fiber material and polymers. On the other
hand, high hydrophobicity of a surface can prevent penetration of
polymer molecules in the sheet, in case high water content of a polymer

dilutes. In this investigation, the extensibility and formability of the
BSKP fiber network was increased by an average of 4% with the
polymer additions via the use of the spray. Fiber network properties
dominate the mechanical behavior of the structure unless a polymer
penetrates into fiber network or forms a uniform phase. The investigated agro-based carbohydrate polymer, starch acetate, could not
improve extensibility or strength of the BSKP sheet, because of the poor
penetration into the fiber network.
The results show that extensibility of the fiber network has an optimum temperature, which is emphasized when polymers are present.
Polymer softening and adhesion play a role in the extensibility.

418


Carbohydrate Polymers 186 (2018) 411–419

J. Kouko et al.

8(1), 472–486.
Zhang, Y., Li, H., Li, X., Gibril, M. E., & d Yu, M. (2014). Chemical modification of cellulose by in situ reactive extrusion in ionic liquid. Carbohydrate Polymers, 99,
126–131. />Östlund, M., Borodulina, S., & Östlund, S. (2011). Influence of paperboard structure and
processing conditions on forming of complex paperboard structures. Packaging
Technology and Science, 24(6), 331–341. />
Vishtal, A. (2015). Formability of paper and its improvement. Doctoral thesis. Tampere
University of Technology. VTT Technical Research Centre of Finland VTT Science 94
108 p.
Waterhouse, J. (1976). The deformation characteristics of polymer reinforced fiber networks. Tappi, 59(7), 106–109.
Zeng, X., Vishtal, A., Retulainen, E., Sivonen, E., & Fu, S. (2013). The elongation potential
of paper—how should fibres be deformed to make paper extensible? BioResources,

419