Synthesis of galactoglucomannan-based latex via emulsion polymerization

Carbohydrate Polymers 291 (2022) 119565

Contents lists available at ScienceDirect

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

Synthesis of galactoglucomannan-based latex via emulsion polymerization
Qiwen Yong a, b, *, Jiayun Xu b, c, Luyao Wang b, Teija Tirri b, Hejun Gao a, Yunwen Liao a,
Martti Toivakka b, **, Chunlin Xu b, **
a

Institute of Applied Chemistry, Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, College of Chemistry and Chemical Engineering, China
West Normal University, Nanchong 637009, China
b
Laboratory of Natural Materials Technology, Åbo Akademi University, Turku 20500, Finland
c
Tianjin Key Laboratory of Pulp and Paper, Tianjin University of Science and Technology, Tianjin 300457, China

A R T I C L E I N F O

A B S T R A C T

Keywords:
Hemicellulose
Galactoglucomannan
Bio-based latex
Packaging application
Etherification
Semi-continuous emulsion polymerization

This is the first time to report a facile strategy to fabricate galactoglucomannan-based latex with highly trans­
parent, hydrophobic and flexible characteristics by combining etherification with subsequent emulsion poly­
merization. The allylated galactoglucomannans (A-GGM) and galactoglucomannan-based latexes (GGM-L) were
prepared and their chemical structure, substitution degree, molecular weight, conversion rate, particle size and
zeta potential were characterized by ATR-FTIR, 1HNMR, quantitative 13CNMR, HP-SEC, HPLC and zeta-sizer
nanometer analyzer, respectively. Furthermore, the effects of substitution degree on film surface roughness
and homogeneity, water vapor permeability (WVP) and thermal stability were evaluated by AFM, SEM, WVP and
TGA, respectively. The optimal GGM-L film exhibited 91.3% transmittance and 0.43% haze, 117◦ water contact
angle, 31.2% elongation at break and 30.9 MPa ultimate tensile stress. The bio-based content of the GGM-L may
reach about 99 wt%, which provides a promising avenue for polyolefin-based latex replacement for paper and
paperboard applications.

1. Introduction
Synthetic polymers or plastics are widely used for packaging,
coating, paints, and other applications in our daily life, and are accu­
mulated in the landfills, oceans, waterways, and other natural envi­
ronment, being a severe burden to the nature. The replacement of fossilbased resources by renewable and sustainable biomass is the basic idea
behind the “Bio-economy”, which is gaining popularity in a wide range of
industries at present (Lehtonen et al., 2016; Mankar, Pandey, Modak, &
Pant, 2021). Agricultural- and forest-based biomass, such as cellulose,
lignin and hemicelluloses, have attained widespread concern owing to
their abundant availability, bioactivity, biocompatibility and biode­
gradability. Moreover, they do not compete with food supply in the
production of biopolymers (Chio, Sain, & Qin, 2019; Schatz & Lecom­
mandoux, 2010).
Hemicellulose (HC), the most abundant plant polysaccharide next to
cellulose, accounts for 20–35% of the total wood mass, yet scarcely
being exploited due to lack of economically feasible applications. Use of

hemicellulose-based films and coatings as green packaging materials to

replace petroleum-based synthetic plastics can contribute to sustainable
development. In the soft wood, the main hemicellulose type is O-acetylgalactoglucomannan (GGM), which consists of a main backbone of
randomly distributed (1 → 4)-linked β-D-mannopyranosyl and (1 → 4)linked β-D-glucopyranosyl units, while (1 → 6)-linked α-D-galactopyr­
anosyl units are attached to mannose units and mannose units are
partially acetylated at C2 and C3 positions (Mikkonen, 2020; Zhao,
Mikkonen, Kilpelainen, & Lehtonen, 2020). GGM has high water solu­
bility and is film-forming to some degree (Battista, Zuliani, Rizzioli,
Fusco, & Bolzonella, 2021). The low oxygen permeability of GGM films
make them promising candidates for substitution of traditional pack­
aging barrier materials, such as aluminum foil, PE, PP, PVDC or EVOH
(Nechita & Roman, 2020). However, the pristine GGM films have some
inherent drawbacks due to GGM's high number of hydroxyl groups,
strong intermolecular hydrogen bonds and low molecular weights.
These result in, for instance, brittleness, low flexibility, high hydrophi­
licity, water/moisture sensitivity, and poor mechanical and thermal

* Correspondence to: Q. Yong, Institute of Applied Chemistry, Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, College of Chemistry
and Chemical Engineering, China West Normal University, Nanchong 637009, China.
** Corresponding authors.
E-mail addresses: (Q. Yong), (M. Toivakka), (C. Xu).
/>Received 17 January 2022; Received in revised form 27 April 2022; Accepted 29 April 2022
Available online 4 May 2022
0144-8617/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license ( />

Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

properties (Khwaldia, Arab-Tehrany, & Desobry, 2010).
Physical and chemical modifications of GGM by either surface or

bulk modification are considered as effective ways to overcome the
above-mentioned short-comings (Peng, Du, & Zhong, 2019). The phys­
ical blending of the hemicellulose with plasticizers including glycerol,
sorbitol xylitol, or emulsifiers like sucrose ester, palmitic acid, or poly­
vinyl alcohol (PVOH) can greatly enhance the film-formation, reduce
the brittleness and provide flexibility for GGM-based composite films.
However, the composite films are hydrophilic and sensitive to water and
ănen,
moisture, as reported previously (Mikkonen, Heikkilă
a, Helen, Hyvo
& Tenkanen, 2010). Various chemical modifications including oxida­
tion, reduction, etherification, esterification, ionization, amination,
fluorination, acetylation, graft polymerization and crosslinking have
been developed to improve the hydrophobicity, tensile property, ther­
mal stability and film-formation of GGM bio-based film materials
(Kisonen et al., 2014; Liu et al., 2019; Markstedt, Xu, Liu, Xu, & Gate­
nholm, 2017; Yi, Xu, Wang, Huang, & Wang, 2020; Zoldners & Kiseleva,
2013). Nevertheless, the film flexibility/stretchability of modified GGMbased films are still hardly comparable to those of petroleum-derived
synthetic polymer materials. For example, Anette Larsson et al. (Har­
delin, Bernin, Borjesson, Strom, & Larsson, 2020) recently reported
etherified GGM films with high hydrophobicity but limited flexibility
with tensile strain at break less than 9%. Frederic Becquart et al. (Farhat
et al., 2018) introduced a long hydrophobic polycaprolactone (PCL) side
chains to hemicellulose by ring-opening graft polymerization, providing
a significant thermoplastic and hydrophobic alteration. It not only
increased the molecular weight of hemicellulose-based film, but it also
broke hydrogen bonds between the molecular chains of hemicellulose,
thereby improving film-forming ability and stretchability of
hemicellulose-based product. However, this graft polymerization was
carried out in organic solvent, which is undesirable from the green

chemistry point of view. Therefore, developing sustainable
hemicellulose-based packaging film materials with good transparency,
high hydrophobicity and tensile property, excellent barrier property and
thermal stability is still a considerable challenge.
Emulsion polymerization is an effective way to achieve free radical
polymerization of water-insoluble organic monomers and water-soluble
substances in an aqueous medium system with the help of emulsifiers. In
light of our previous research results (Kisonen et al., 2014; Yong &
Liang, 2019), we anticipate that designing GGM-based biopolymer
grafted by long-chain hydrophobic organic monomers may achieve both
hydrophobicity and flexibility, due to the introduction of hydrophobic
macromolecular side-chains and the increase of molecular weight of
GGM-based latex. In this paper, we first used allyl glycidyl ether (AGE)
to etherify the native GGM between epoxy groups of AGE and hydroxyl
groups of GGM, as ether bond is more stable than the ester and amide
bonds, of which the latter are more easily hydrolyzed under acid/alkali
conditions (Laine et al., 2013). Then a semi-continuous emulsion poly­
merization method was used to copolymerize between allylated GGM
and hydrophobic acrylate monomers n-butyl acrylate (n-BA) under the
assistance of emulsifier sodium dodecyl sulfate (SDS) and initiator
ammonium persulfate (APS). It should be noted that n-butyl acrylate can
be obtained from biomass resources based on the bio-based n-butanol
fermented from glucose and the bio-based acrylic acid converted from
ăkkilă
lactic acid (Garcớa, Pa
a, Ojamo, Muurinen, & Keiski, 2011; Niesbach,
Fink, Lutze, & G´
orak, 2015; Yang et al., 2021). The bio-based content
can reach up to 99% by weight, calculated from the solid matter of the
latex, when the n-BA is obtained from biobased raw-materials. In gen­

eral, a series of sustainable GGM-based latexes with different substitu­
tion degrees were prepared successfully. The films exhibited good
transparency, high hydrophobicity and flexibility, excellent barrier
property and thermal stability, indicating that there is a promising
substitution of conventional petroleum-based synthetic polymer mate­
rials in paper and paperboard coatings.

2. Experimental procedure
2.1. Materials
Pressurized hot-water extracted galactoglucomannan from Norway
spruce was obtained as a solution with concentration of 10 wt% from the
Finnish Forest Research Institute Metla. Allyl glycidyl ether (98%), so­
dium hydroxide (NaOH, AR), sodium dodecyl sulfate (AR), ammonium
persulfate (95%), and n-Butyl acrylate (AR) were provided by SigmaAldrich. Hydrochloric acid (HCl), ethanol (98%), acetone (99%) and
methyl tert-butyl ether (MTBE, AR) were purchased from Altia
Industrial.
2.2. Methods
2.2.1. GGM extraction and characterization
The native GGM dispersion was precipitated by industrial grade
ethanol at a 2:8 (v/v) water-ethanol ratio. Glass fiber filter was applied
for vacuum filtration. Then the filter cake was redissolved into water,
and the ethanol precipitation and filtration were repeated. After the
third filtration, the filter cake was washed twice consecutively by
flushing with pure ethanol, acetone and MTBE. The purified GGM
filtrate was dried overnight at ambient temperature, and further dried in
a vacuum desiccator at 40 ◦ C for 48 h to remove the residual MTBE. The
purified GGM had a number average molecular weight (Mn) of 8.31 ×
103 g/mol and a weight average molecular weight (Mw) of 19.80 × 103
g/mol (polydispersity ~2.38), as measured by a high-performance size
exclusion chromatography (HPSEC). The sugar unit ratio of GGM was

analyzed using a gas chromatography (GC), and it was at 0.67:1:3.26
(Galactose: Glucose: Mannose). The composition of GGM characterized
by 2D HSQC-NMR was shown in Fig. S6 and 13C–1H correlation signals
were listed in Table S5.
2.2.2. Synthesis of allylated GGM dispersions
The GGM was modified by an etherification reaction with a bifunc­
tional monomer or crosslinker AGE at three molar ratios (see Table S1)
between moles of anhyrohexose unit of GGM and moles of AGE used (i.
e., 1:0.25, 1:0.5, 1:1, 1:2 and 1:3, for the five reactions) (Hardelin et al.,
2020; Laine et al., 2013). Specifically, 10 g of vacuum-oven dried GGM
was mixed with 200 mL of deionized water and 2% of NaOH (based on
the total weight of the reaction system) was added into a three-necked
flask. The reaction system was equipped with a condenser and pro­
tected by a flowing N2 atmosphere, and was also stirred by a magnetic
stirrer at 600 rpm. The temperature raised to 60 ◦ C and the AGE was
added dropwise by a dropping funnel. The 10 h reaction time
commenced once all the AGE was added into the reaction flask. After­
– C double bonds was cooled and
wards, the modified GGM containing C–
neutralized with HCl. Finally, the allylated GGM (A-GGM) were further
dialyzed with membranes with a molecular weight cut-off of 12– 14 ×
103 g/mol for at least 72 h.
1
H NMR (500 MHz, DMSO‑d6): δ (ppm) = 2.0 (H-1), 4.0 (H-4), 4.5
(H-3, α), 5.1, 5.3 (H-6), 5.3 (H-3, β), 5.9 (H-5);
13
C NMR (125 MHz, DMSO‑d6): δ (ppm) = 21.5 (C-1′ ), 101.2 (C-3′ ),
116.0 (C-6′ ), 135.2 (C-5′ ), 170.1 (C-2′ ).
2.2.3. Synthesis of GGM-based latexes
GGM-based latexes (GGM-L) were synthesized by semi-continuous

emulsion polymerization technique. The detailed recipes are in
Table S2. In brief, 5 g of above allylated GGM was concentrated into a
volume of 100 mL by a rotary evaporator and then was transferred into a
250 mL three-necked round-bottom flask equipped with a condenser, a
peristaltic pump with injection needle and a N2 circulation device. 1% of
emulsifier SDS (based on the weight of n-BA) was added to the flask and
dissolved thoroughly. The reaction system was heated to 80 ◦ C and
0.75% of APS (based on the weight of n-BA) was added to the flask.
Then, 5 g of n-BA was added dropwise into the reaction system by the
2


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Fig. 1. (a) Molecular structural formula of GGM and its etherification principle; (b) 1H NMR and (c) quantitative
specimens; (d) ATR-FTIR and (e) partial enlarged ATR-FTIR spectra of native GGM and A-GGM specimens.

3

13

C NMR spectra of native GGM and A-GGM


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

peristaltic pump during 30 min. The reaction temperature was kept at
80 ◦ C for 3 h with the stirring speed of 600 rpm. Finally, another 0.25%
of APS initiator was added, and the reaction continued for 1 h to ensure
that the remaining monomers had completely reacted. The prepared
GGM-based latexes were stored in a cold room (4 ◦ C) for further use. The
solid content of the GGM-L latexes was about 12% and the apparent
viscosity is reported in Fig. S1.
1
H NMR (500 MHz, DMSO‑d6): δ (ppm) = 0.9 (H-1), 1.3 (H-2), 1.5
(H-3), 2.4 (H-6), 5.1 (H-7), 5.3 (H-7), 5.9 (H-8).
13
C NMR (125 MHz, DMSO‑d6): δ (ppm) = 14.4 (C-1′ ), 18.9 (C-2′ ),
28.0 (C-3′ ), 35.7 (C-6′ ), 65.9 (C-4′ ), 116.0 (C-8′ ), 135.2 (C-7′ ), 174.8 (C5′ ).

Table 1
Allylation degrees and molecular weights of GGM and A-GGM.
Mw[×103 g/mol]b

Mn[×103 g/mol]b

PDIb

GGM
A1-GGM
A2-GGM
A3-GGM
A4-GGM
A5-GGM

–

0.04
0.13
0.30
0.51
0.83

19.80
22.04
23.53
24.16
30.36
32.58

8.31
7.54
7.86
6.68
8.17
7.90

2.38
2.92
2.99
3.60
3.72
4.12

b

Represents the DSal determined by quantitative 13CNMR.

Measured by HPSEC with DMSO/LiBr as solvent and eluent.

Theoretically, the full substitution degree of GGM is 3, which in­
dicates that all three hydroxyl groups of each anhydrosugar units have
been grafted with AGE molecules. Based on the quantitative 13C NMR
spectra, we can calculate the DSal. With the molar ratio of GGM:AGE
increasing from 1:0.25 to 1:3, the DSal of GGM increased from 0.04 to
0.83. Compared to the DSal values of 1.1, 1.9, and 2.5 reported in
(Hardelin et al., 2020), our substitution degree was lower. This is due to
the low molar ratio of AGE:GGM and the low amount of NaOH used in
our work. Moreover, our aim was to increase the bio-based content. The
detailed DSal data and molecular weight of GGM and A-GGM products
can be found in Table 1. The Mw and polymer dispersity index (PDI) of
pristine GGM are 19.80 × 103 g/mol and 2.38, respectively (Hardelin
et al., 2020). In general, the Mw and PDI of A-GGM increased with
increasing the AGE contents. This can be attributed to the introduction
of more allyl chains into GGM backbones due to the continuously
increasing of DSal.
The comparison of the ATR-FTIR spectra of GGM and allylated GGMs
also identified the attachment of allyl groups onto the gal­
actoglucomannan backbone, as shown in Fig. 1d and e. In terms of un­
modified GGM, the wide band at 3300–3600 cm− 1 represents the free
and H-bonded -OH stretching vibrations of GGM (Liu, Renard, Bureau, &
Le Bourvellec, 2021). The weak band attributed to the deformation vi­
brations of -OH is also displayed at 1631 cm− 1 (Maleki, Edlund, &
–O
Albertsson, 2017). The band at 1725 cm− 1 is attributed to the C–
carbonyl stretching vibrations of the acetyl groups of GGM (Zasadowski,
Yang, Edlund, & Norgren, 2014). With regard to the synthesized ally­
lated GGM, a strong wedge-shaped band at 1643 cm− 1 is assigned to the

– C stretching vibrations and a small band at 810 cm− 1 is associated
C–
with an unsaturated C–H bending vibrations, confirming the successful
etherified modification of AGE onto pristine GGM main chains. More
importantly, the increased band intensity at 1643 cm− 1 also verifies the
increasing content of double bonds attached onto the hemicellulose
backbone, which is in good agreement with the degree of substitution
determined by the use of carbon resonance signals in 13C NMR spectra.

2.3. Characterizations
Detailed information regarding instrumental setups and experi­
mental procedures employed for the characterization of pristine GGM,
A-GGM and GGM-L latexes/ films is provided in Supporting Information.
3. Results and discussion
3.1. Characterization of A-GGM
– C double bonds were grafted on GGM backbone to prepare an
C–
allylated GGM with suitable substitution degree to enable subsequent
free radical copolymerization. This was achieved through etherification
reaction between hydroxyl groups of GGM and epoxy groups of AGE
performed in aqueous NaOH solution. The molecular structural formula
of GGM and its etherification principle are shown in Fig. 1a. Meanwhile,
the chemical structure and substitution degree of allylated GGM were
confirmed by 1H NMR and quantitative 13C NMR spectra before and
after modification, as shown in Fig. 1b and c. Prior to allylation, the
single peak appearing at 2.0 ppm (1) in the 1HNMR spectra is ascribed to
the methyl protons of pendant acetyl moieties of GGM, and the peaks at
3.1–5.3 ppm are corresponding to the protons of carbohydrate backbone
(Chadni, Grimi, Bals, Ziegler-Devin, & Brosse, 2019). After allylation,
the split peaks at 5.1 ppm and 5.3 ppm (6), as well as newly emerging

signal at 5.9 ppm (5) originate from three unsaturated vinylene protons.
Moreover, the new resonance signal at 4.0 ppm (4) is attributed to the
methylene protons connected to the vinyl group (Leppanen et al., 2014).
All these results verify that the pendant allylated groups were success­
fully grafted onto the hemicellulose backbone. With regard to the
quantitative 13C NMR spectra, the resonances at 21.5 (1′ ) and 170.1 ppm
(2′ ) can be separately assigned to methyl and carbonyl groups of acetyl
moieties, which are diminished in the allylated GGM due to the hy­
drolysis reaction in alkali solution (Hardelin et al., 2020). In contrast,
the new peaks at 116.0 (6′ ) and 135.2 ppm (5′ ) are ascribed to the vinyl
groups of allylated GGM, which indicates that the success of ether­
ification between the hydroxyl groups of native GGM and epoxy groups
of AGE.
The substitution degree of allylation (DSal) was calculated from the
– C double bond peaks at 116.0 and 135.2 ppm
relative integrals of the C–
against the constant C1 peak at 101 ppm (3′ ) on quantitative 13C NMR
spectra, according to Eq. (1):
(I116 + I135 )/2
I101

DSala

a

2.2.4. Preparation of GGM-based films
Films from the pristine GGM, A-GGM and GGM-based latexes were
prepared in 10 cm diameter petri dishes with covers. Known amounts of
dispersion in petri dishes were dried at constant temperature of 23 ◦ C
and humidity of 50% for 3 days. The thickness of the dried films was 80

± 5 μm based on ten measurements using a Lorentzen and Wettre
Micrometer. The prepared GGM-L films were kept in sealed bags and
stored in a vacuum desiccator for further analysis.

DSal =

Samples

3.2. Characterization of GGM-L
– C double
In the synthesis of GGM-based latexes, the unsaturated C–
bonds of AGE grafted on the hemicellulose backbone were used as re­
action sites of the free-radical copolymerization of n-BA in semicontinuous emulsion polymerization process, which is sketched in
Fig. 2a. The representative ATR-FTIR spectra of A-GGM and GGM-L
products are compared in Fig. 2b. It can be seen that the newly
emerging sharp band at 1735 cm− 1 is attributed to the carbonyl
stretching vibrations from the ester groups of n-BA after copolymeriza­
tion. The emulsion polymerization also can be identified by the disap­
– C vibration at 1643 cm− 1, revealing that most of the
pearance of C–
– C double groups from either allylated GGM or n-BA are
unsaturated C–
consumed in the polymerization reaction.
1
H NMR and 13C NMR spectra were further to confirm the chemical
structure of GGM-L specimens. In the 1H NMR spectra of A-GGM and
GGM-L (Fig. 2c), The newly-emerged resonances at 0.9 (1), 1.3 (2), 1.5
(3) and 2.4 ppm (6) were originated from the n-BA monomer, especially
the resonance at 2.4 ppm represented the hydrogen atoms connecting to

(1)

4


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Fig. 2. (a) Schematic outline of the synthesis of GGM-L latexes; (b) ATR-FTIR and (c) 1H NMR spectra of typical A-GGM and GGM-L specimens; (d) Hemicellulose
biomass content of GGM-L specimens; (e) 13C NMR spectra of typical A-GGM and GGM-L specimens.
5


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

ester group after copolymerization. The extensive intensity decreased in
resonances at 5.9 (7), 5.3 and 5.1 ppm (8), as well as the resonance at
4.0 ppm corresponding to vinyl groups and its neighbored methylene
–C
groups respectively, demonstrating that nearly all the unsaturated C–
double bonds in the reactive system have participated in the copoly­
merization reaction. In case of 13C NMR spectra of A-GGM and GGM-L
(Fig. 2e), the new resonances from 13C NMR spectrum of GGM-L at
174.8 (5′ ), 65.9 (4′ ), 35.7 (6′ ), 28.0 (3′ ), 18.9 (2′ ) and 14.4 ppm (1′ ) were
from n-BA, which demonstrates that n-BA has introduced successfully
into the system. More convincingly, the disappearance of resonances at
135.2 (7′ ) and 116.0 ppm (8′ ) from 13C NMR spectrum of A-GGM in­

dicates that the double bonds of A-GGM had reacted completely with nBA monomer. The resonance at 130.1 ppm in 13C NMR spectrum of
GGM-L shows that there was still a small amount of unreacted n-BA

Table 2
Molecular weights of GGM-L and polymerization degrees of poly(n-butyl
acrylate).
Samples

Mw[×104 g/mol]a

Mn[×104 g/mol]a

PDIa

DPb

GGM-L1
GGM-L2
GGM-L3
GGM-L4
GGM-L5

2.43
2.75
3.46
7.73
22.34

1.48
1.28

1.34
2.72
9.74

1.64
2.15
2.58
2.84
2.30

17.63
30.97
81.45
366.23
1488.80

a
b

Measured by HPSEC with DMSO/LiBr as solvent and eluent.
Estimated polymerization degree (DP) = (Mw (GGM-L) – Mw (A-GGM))/128.17.

Fig. 3. (a) Photos of the prepared A-GGM and GGM-L dispersions; (b) and (c) are particle size distributions of the A-GGM and GGM-L specimens, respectively; (d) and
(e) are Z-average nanoparticle sizes and zeta potentials of A-GGM dispersions and GGM-L latexes, separately.
6


Q. Yong et al.

7

Carbohydrate Polymers 291 (2022) 119565

Fig. 4. (a) MilliQ water and ethylene glycol contact angles of GGM-L films; (b) Polarity force, dispersion force and surface free energy of GGM-L films; (c) AFM images and surface roughness of GGM-L1, GGM-L3 and
GGM-L5 films, respectively. The AFM images and surface roughness of GGM-L2 and GGM-L4 films are in Fig. S4.


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

residual monomer in the system. This result can be verified from Fig. S7,
which displays the reactive conversion rate of all GGM-L samples.
Compared to other GGM-L products, the reactive conversion rate of
GGM-L1 was slightly lower. Notably all of GGM-L specimens demon­
strate that the reactive conversion rates have exceeded 99%.
It is well accepted that biomass content is an important indicator in
developing biomass-based products. On the one hand, it will reduce the
usage of fossil fuels in starting materials. On the other hand, the more
biomass materials we use, the easier the products based on them can
biodegrade. Fig. 2d demonstrates the biomass content in the synthesized
bio-based GGM-L products. Obviously, with the increase of AGE content,
the biomass content slightly decreased from 49.1% to 38.7%. This is due
to the fact that the DSal increased with an increase in AGE content,
indicating that more AGE was consumed and grafted on GGM molecular
structure. In general, the usage of GGM which accounts for nearly 45 wt
% of total mass, reveals a positive beginning in preparing highly hy­
drophobic hemicellulose-based coatings. If we use commercially biobased n-butyl acrylate to substitute the fossil-based one, the biomass
content of the GGM-L latexes can reach up to 99 wt%, which shows a
tremendous perspective for green and sustainable future.
Table 2 shows the average molecular weights of GGM-L latexes and

the estimated degrees of polymerization (DP) of poly(n-butyl acrylate).
When the allylated substitution degrees (DSal) of A-GGM samples
increased from 0.04 to 0.83, the weight average molecular weight (Mw)
of corresponding GGM-L samples increased from 2.43 × 104 g/mol to
22.34 × 104 g/mol and the number average molecular weight (Mn)
increased from 1.48 × 104 g/mol to 9.74 × 104 g/mol. Since the length
of side chains on the GGM backbone is of critical importance for the
properties of GGM-L, the average polymerization degrees of poly(nbutyl acrylate) attached on GGM backbone were estimated. The DP
values of GGM-L1 and GGM-L2 were 17.63 and 30.97, respectively,
which indicates that the grafted poly(n-butyl acrylate) chains were not
too long, and therefore it had less impact on properties of GGM-L. This
may be attributed to the low DSal of A1-GGM and A2-GGM samples. With
an increase of DSal from 0.30 to 0.51, the DP values of GGM-L increased
from 81.45 to 366.23. With the long molecular chains of poly(n-butyl
acrylate) attached on GGM backbone, GGM-L3 and GGM-L4 films
showed enhanced hydrophobicity and stretchability compared with the
native GGM. In case of GGM-L5 sample, it had the highest Mw of 22.34 ×
104 g/mol and Mn of 9.74 × 104 g/mol, as well as the highest DP of
1488.80 among all GGM-L samples. It can be inferred that there were
many branched poly(n-butyl acrylate) chains crosslinked each other due
to the high DSal (0.83) of A5-GGM sample, leading to the sharp increase
in molecular weight and degree of polymerization of GGM-L5 sample.
Therefore, we can conclude that the adjustment of DSal in the ether­
ification step is critical to prepare the GGM-L latexes with improved
hydrophobicity and stretchability, and to avoid the fabrication of gels of
GGM-L.

weighted average hydrodynamic diameter (Z-average) of GGM is mainly
dependent on the GGM raw materials we used. All of the A-GGM latexes
display the particle size of around 100 nm. The specific Z-average par­

ticle size data of A-GGM and GGM-L are in Fig. 3d. However, as shown in
Fig. 3c, the Z-average particle size of all GGM-L emulsions increases
considerably when compared to the A-GGM dispersions, and the particle
size increases with increasing the DSal of A-GGM. Furthermore, the
width of particle size distribution also increases also with the increase of
substitution degree of the A-GGM. These results are attributed to the
higher the degree of substitution, the higher number of allylated groups
grafted on the hemicellulose backbone. Hence, the etherified GGM
products were able to provide more reactive sites. With polymerization
taking place within in micelles where A-GGM and monomer are solu­
– C double
bilized within clusters of SDS surfactant molecules, the C–
bonds from both A-GGM and n-BA are initiated by a free radical, origi­
nating from the decomposition of the initiator APS, and can in turn add
– C double bond to form large molecules of repeating
on to another C–
units. The micelle particles obviously increase in size during the poly­
merization. Particle stability is maintained by further adsorption of SDS
molecules at the particle surface.
Zeta potential (ζ-potential) is caused by the net electrical charge
contained within the region bounded by the hydrodynamic slipping
plane. The magnitude of the ζ-potential reveals the degree of electro­
static repulsion between adjacent charged particles in a dispersion.
Therefore, colloids with high ζ-potential (negative or positive) are
electrostatically stabilized while colloids with low ζ-potential tend to
coagulate or flocculate. As we can see from Fig. 3e, the absolute value of
ζ-potential of all GGM-L latexes is much higher than that of all the AGGM dispersions, due to the introduction of SDS surfactant in the second
stage of emulsion polymerization. All the GGM-L specimens show a
relatively high absolute ζ-potential value of over 25 mV, indicating that
the synthesized GGM-L emulsions have a moderate electrostatic stability

to avoid coagulation. To further increase the ζ-potential and thereby also
the emulsion stability, acrylic acid could be added in the emulsion
polymerization process. In general, the resultant GGM-L latexes having
particle sizes of between 80 and 210 nm are suitable for technical use,
for example, in paper and paperboard barrier coating applications.
3.4. Film hydrophobicity and surface energy
The hydrophobicity of a coating is a crucial parameter for packaging
materials(Cunha & Gandini, 2010). The results in Fig. 4a show that
contact angles for both milliQ water and ethylene glycol increase with
increasing the DSal of A-GGM from GGM-L1 to GGM-L4 films. GGM-L5
film shows a small decrease in the contact angles compared to GGM-L4
film. For instance, GGM-L1 (DSal 0.04) shows an average water contact
angle of 62.9◦ which is still hydrophilic, whereas GGM-L4 (DSal 0.51)
film demonstrate the water contact angle up to 117.2◦ , indicating an
high hydrophobicity. In the case of GGM-L5 specimen, the decrease of
milliQ water and ethylene glycol contact angles may be due to the
surface roughness of the GGM-L5 film (Farhat et al., 2018). This can be
verified by AFM results shown in Fig. 4c and Table S3. With an increase
of DSal values, the GGM-L film surfaces became much rougher. For
example, GGM-L1 film had a film roughness of 5.25 nm of Ra and 6.92
nm of Rq, while GGM-L3 film possessed a surface roughness of 15.5 nm
of Ra and 21.4 nm of Rq, respectively. GGM-L5 film declared a DSal of
0.83, which demonstrated the highest roughness up to 54.9 nm of Ra and
67.9 nm of Rq respectively. Therefore, the heteropolymer of A-GGM with
a greater high DSal is copolymerized with n-BA monomer to form crosslinking and interpenetrating network between the molecular chains,
leading to the gel formation of GGM-L5 product, which in turn shows a
greater surface roughness on its film surface (Wenzel, 2002).
The Owens-Wendt method (also known as the Kælble-Owens-Wendt
method was used for calculating the surface energies of a series of GGML films and for resolving the surface energy into contributions from
polarity and dispersion forces (Rudawska & Jacniacka, 2009; Wei, Zeng,

3.3. Particle size and zeta potential
In the majority of coating applications, particle size and zeta po­
tential are highly important and readily measurable indicators that
determine the properties of a polymer dispersion, such as flow behavior
or stability. In order to eliminate the impact of multiple scattering of the
laser light and particle interactions, which influence diffusion, all sam­
ples were diluted to 0.01% w/w before the measurements. As we can see
from Fig. 3a, all A-GGM and GGM-L specimens are brownish yellow in
color, but the color of GGM-L latexes is slightly lighter than that of AGGM dispersions after emulsion polymerization. Also, the GGM-L la­
texes are more viscous with increasing the molar ratio of GGM:AGE,
which can be verified from data in Fig. S1. This is due to the higher
substitution degree that may result in a higher level of cross-linking
between the macromolecular chains of GGM during the copolymeriza­
tion with n-BA. In Fig. 3b, the etherification modification shows little
influence on the particle size and particle size distribution. The intensity
8


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Fig. 5. (a) Digital photos of GGM-L films covering on logos with copyright permission from Åbo Akademi University; (b) Optical transmittance and (c) transmission
haze of GGM-L films.

& Yong, 2021). The polarity force is caused by the sum of polar,
hydrogen, inductive and acid-base interactions. The London dispersion
force arises from instantaneous dipoles produced by the motion of
electrons within the molecule. The combination of Eqs. (2) and (3) allow

one to determine the surface free energy (SFE) and its components:

accounts for a large proportion of total SFE, indicating that it contributes
to a major part of the strength of GGM-L bio-based films.

γ S = γ dS + γ pS

(2)

)1/2
)1/2
(
(
γ L (1 + cosθ) = 2 γ dS γdL
+ 2 γ PS γ PL

(3)

Fig. 5a shows digital photographs of the GGM-L films. In general, the
films show the color of light brown, but one can clearly see the logos
beneath the films. Fig. 5b shows the measured light transmittance of the
films as a function of wavelength. All the GGM-L films were quite
transparent in the visible light range, especially at high wavelength.
With increasing the DSal from 0.04 to 0.51, the light transmittance of
GGM-L films increased slightly from 86.1% of GGM-L1 to the highest
– C bond
value of 91.3% for GGM-L4, which might be influenced by C–
– C bonds either from A-GGM or n-butyl
conversion rate. Unreacted C–
monomer reduces film transparency. The slight decrease of GGM-L5 film

transparency was not only related to conversion rate, but also influenced
by its gel characteristic. All the prepared GGM-L films show a steep drop
of transmittance in the ultraviolet light range, i. e., below 400 nm,
indicating a superior block of UV light transmittance. This result can be
of benefit for protecting products in packaging coatings as the most
pronounced detrimental effects of light are usually induced by ultravi­
olet light (Mikkonen et al., 2010; Song, Xu, Li, Chen, & Xu, 2021).
Compared to the small change of light transmittance ranging from
86.1% to 91.3% at 550 nm, there are higher variations in the optical
haze of the GGM-L films (in Fig. 5c). Haze can be used to quantify the
percentage of the forward light scattering (wide angle scattering), and is
experimentally expressed as Eq. (4): (Zhu et al., 2013)

3.5. Film transparency and haze

where γdS denotes the component of SFE due to dispersion force, γpS de­
notes the component of SFE related to polarity force, γ S represents the
total SFE of a solid surface. The γL, γ dLand γ PL values of two measured
liquids (milliQ water and ethylene glycol) can be obtained from Table S4
in supplementary materials.
As can be seen from Fig. 4b, with the increase of DSal of A-GGM, the
surface energy decreased from 40.6 mJ m− 2 for GGM-L1 film to 10.3 mJ
m− 2 for GGM-L4 film. We should always bear in mind that the SFE below
30 mJ m− 2 for a coating means superior hydrophobicity of polymers
such as polypropylene (Yong et al., 2016). The low SFE values of GGML3, GGM-L4 and GGM-L5 products confirmed the success of hydrophobic
modification of the hydrophilic GGM raw material. In the case of SFE
components, the values of polarity force decreased from 19.4 mJ m− 2 to
0.4 mJ m− 2 in GGM-L1 ~ GGM-L4 samples, indicating that the film
surface became less polarity due to the extension of polymer chains
caused by the free radical polymerization with n-BA monomer.

Compared with the polarity component, London dispersion force
9


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Fig. 6. (a) Diagram of spline preparation and clamping device; (b) Mechanical stress-strain curves of bio-based GGM-L films; (c) Elastic modulus and toughness of
bio-based GGM-L films; (d) and (e) represent cross-section images of GGM-L4 and GGM-L5 films separately fractured by liquid Nitrogen. The cross-sections of GGM-L1,
GGM-L2 and GGM-L3 films can be seen in Fig. S5.

(
Haze (%) =

)
T4 T3
−
× 100%
T2 T1

(4)

Table 3
The mechanical properties of the GGM-L films.

where T1, T2, T3 and T4 are defined in the Fig. S3. With the increase of
AGE dosage, the transmission haze of the GGM-L films at the wavelength
of 550 nm demonstrates an obvious decreased trend, and the haze values
are 3.94%, 2.05%, 0.96%, 0.43% and 1.26%, respectively. The result is

more likely due to the higher substitution degree of allylated groups
leading to a high polymerization density that reduces the light scattering
(Tong, Chen, Tian, & He, 2020). In the case of the GGM-L5 film, it was
found that the haze slightly increased to 1.26%, which might be caused
by the surface roughness. Besides, the transmission haze values of these
GGM-L films are well matched with the visual effect of Fig. 5a, which
can be directly judged by clear or cloudy appearance. The results indi­
cate that the GGM-L films, especially for GGM-L4 film, possess high
optical transparency and relatively low transmission haze, which are
applicable for packaging barrier materials.

Sample

Elongation at
break (%)a, b

Tensile stress at
maximum load
(MPa)a, b

Elastic
modulus
(MPa)

Toughness
(MJ m− 3)

GGML1
GGML2
GGML3

GGML4
GGML5

8.5 (2)

20.6 (2)

1265.1

157.4

14.4 (4)

28.0 (2)

1321.9

350.6

22.1 (3)

27.8 (1)

1108.1

523.2

31.2 (4)

30.9 (1)

882.3

701.5

3.6 (3)

18.5 (3)

922.0

56.2

a
b

All samples were tested in triplicate.
Standard deviations are presented in parentheses.

films are commonly very brittle (Huang, Zhong, Zhang, & Cai, 2017).
The tensile strains at break of most reported natural polymer-based films
are below 10% (Hardelin et al., 2020; Huang et al., 2017; Qi, Chang, &
Zhang, 2009). Fig. 6a displays the spline shape and clamping device that
were used for tensile test. Fig. 6b and c shows the variation in the

3.6. Film flexibility and uniformity
It is well established that films from natural biopolymers such as
cellulose and hemicellulose, are difficult to form on their own and the
10

Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Table 4
The WVP properties of chemically modified hemicellulose-based films.
Material

Chemical modification

Physical blend

Water vapor
permeability
[g • mm/m2 • d •
kPa]

Reference

Galactoglucomannan

–

–

(Mikkonen et al., 2010)

Galactoglucomannan/PVOH

–

25% PVOH

Galactoglucomannan/CNW

–

Galactoglucomannan/G

Crosslinking with 5% glyoxal

5% Cellulose
nanowhiskers
–

Galactoglucomannan/G/S

Crosslinking with 5% glyoxal

40% Sorbitol

Galactoglucomannan/S

–

40% Sorbitol

Galactoglucomannan/AZC

Crosslinking with 10% AZC

–

Arabinoxylan

–

–

1.7 ± 0.05 (RH 0/
54%)
1.0 ± 0.1 (RH 0/
54%)
2.0 ± 0.3 (RH 0/
54%)
12.0 ± 0.9 (RH 0/
54%)
1.1 ± 0.2 (RH 0/
54%)
2.1 ± 0.09 (RH 0/
54%)
21.8 ± 1.7 (RH 0/
54%)
7.7 (RH 0/52%)

Arabinoxylan
Carboxymethyl xylan

–

Carboxymethylation, DS = 0.36 or 0.58

50% β-glucan
–

9.9 (RH 0/52%)
19.0 ± 2, 38.0 ± 2

Carboxymethyl xylan
Carboxymethyl xylan/G
Hydroxypropyl xylan
Hydroxypropyl xylan/G
Hemicellulose (67.8% Xylan)
Carboxymethyl hemicellulose
GGM (in this study)
GGM-L1

Carboxymethylation, DS = 0.3
Carboxymethylation, DS = 0.3
Hydroxypropylation, DS = 1.1
Hydroxypropylation, DS = 1.1
–
Carboxymethylation, DS = 0.51 or 0.85
–
Etherification, DS = 0.04 and then graft emulsion
polymerization
Same as above, DS = 0.13
Same as above, DS = 0.30
Same as above, DS = 0.51
Same as above, DS = 0.83

–

–
10% glycerol
–
10% glycerol
–
–
–

3.31
1.41 ± 0.32
2.23 ± 0.39
1.75 ± 0.23
5.2
5.9, 6.8
3.0 (RH 0/50%)
0.78 (RH 0/50%)

–

–

GGM-L2
GGM-L3
GGM-L4
GGM-L5
Poly(vinyl alcohol) (PVOH)
Low-density Poly(ethylene)
(LDPE)

–
–
–
–
–

mechanical properties for a series of GGM-based films. Table 3 lists the
data of the maximum elongation at break (ε), ultimate tensile stress (σ),
elastic modulus and toughness of the prepared GGM-L films. As the
amount of AGE increased in first modification step, or the values of DSal
increased, the extensional stress had a slight increase from 20.6 MPa of
GGM-L1 to 30.9 MPa of GGM-L4, but the elongation at break revealed a
pronounced enhanced trend, which was 31.2% for GGM-L4 film and only
8.5% for GGM-L1 film. Generally, GGM-L4 specimen with allylated
substitution degree of 0.51 yielded film with the optimal mechanical
property, i.e., 31.2% of ε and 30.9 MPa of σ. In addition, it possessed the
lowest elastic modulus of 882.3 MPa and the highest toughness of 701.5
MJ m− 3 among all prepared GGM-L samples. The high toughness in­
dicates a high energy requirement for the sample to fracture, provided
by the superior film flexibility of GGM-L4 sample.
Interestingly, as DSal further increased to 0.83, the GGM-L5 film only
exhibited a ε value of 3.6% and a σ value of 18.5 MPa. Even though the
elastic modulus of GGM-L5 sample was not strong, the toughness value
of GGM-L5 sample was only 56.2 MJ m− 3. The result may be explained
by that the branched chains of the hemicellulose were more likely to be
crosslinked and tangled with each other during the second polymeri­
zation stage caused by the high substitution degree of allylated groups.
Therefore, a large amount of cross-linked network structure limited the
movements between and inside macromolecular chains, leading to

lower flexibility of the GGM-L5 film. This also can be verified from
Fig. 6d and e, which show the freeze-fractured cross-sections of GGM-L4
and GGM-L5 films with 250×, 2500× and 5000× enlargements,
respectively. The surface morphology of GGM-L4 cross-sections is quite
smooth and homogeneous, whereas in contrast the GGM-L5 film cross

0.45 (RH 0/50%)
0.41 (RH 0/50%)
0.30 (RH 0/50%)
0.22 (RH 0/50%)
0.5 ± 0.03 (RH 0/
54%)
0.16

(Mikkonen et al., 2010)
(Mikkonen et al., 2010)
(Mikkonen, Heikkilă
a, Willfă
or, &
Tenkanen, 2012)
(Mikkonen et al., 2012)
(Mikkonen et al., 2012)
(Mikkonen, Schmidt, Vesterinen, &
Tenkanen, 2013)
(S
arossy, Tenkanen, Pitkă
anen, Bjerre, &
Plackett, 2013)
(S
arossy et al., 2013)

(Alekhina, Mikkonen, Alen, Tenkanen,
& Sixta, 2014)
(Sousa, 2016)
(Sousa, 2016)
(Sousa, 2016)
(Sousa, 2016)
(Geng et al., 2020)
(Geng et al., 2020)

(Mikkonen et al., 2010)
(Hansen & Plackett, 2008)

sections are relatively uneven and rough. It should be noted that GGM-L5
latex has the unique and highest apparent viscosity among all GGM-L
latexes (in Fig. S1). The high apparent viscosity tends to affect the
fluidity of latex, leading to an increase of surface roughness during the
film formation. This result suggests that excessive cross-linking between
allylated GGM and monomer due to higher DSal might lead to instability
within the emulsion system.
3.7. Water vapor permeability
Oxygen and water vapor barrier properties are considered crucial
with respect to many packaging materials, notably for food packaging
materials (Nechita & Roman, 2020). Hemicellulose has an intrinsic
property of low oxygen permeability due to its specific molecular
structure and intermolecular interactions, which makes it a prospective
application in barrier coatings and films (Stark & Matuana, 2021).
However, the disadvantages of moisture resistance and high water vapor
permeability discourage its use in packaging films and coatings. The
setup for WVP test can be seen in Fig. S2. As can be seen in Table 4, the
reported water vapor permeability (WVP) of HC and HC-based modified

films in literature data ranged from 1.0 to 38.0 g mm/m2 d kPa. The
WVP of the pure GGM film presented in this study is 3.0 g mm/m2 d kPa
at 50% relative humidity and at 23 ± 2 ◦ C. This high permeability is
mainly ascribed to the hydrophilic nature of GGM biopolymer with high
hydration capacity. After modification by etherification and emulsion
polymerization, the WVP values of the GGM-L films produced in the
present research greatly decreased, being about 10 times lower than for
the neat GGM film. The values were similar to the WVP value of Poly
11


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Fig. 7. (a) TGA and (b) DTG curves of the unmodified GGM and modified GGM-based films.

(vinyl alcohol) (PVOH) (0.5 g mm/m2 d kPa), and quite close to the lowdensity polyethylene (LDPE) (0.16 g mm/m2 d kPa). Increasing the
substitution degree of the etherification reaction from DS = 0.04 to DS
= 0.83 showed a positive influence on the moisture barrier properties of
the GGM-L films. All these results can be explained by the breakages of
hydrated hydrogen bond through etherification modification and by the
side chain extensions of GGM through emulsion polymerization. The
higher degree of substitution of the GGM contributed to a higher number
– C double bond reaction sites for allylated GGM, which were pro­
of C–
vided for subsequent emulsion polymerization with monomer, leading
to more macromolecular branches and higher molecular weights.

polymerization process, the thermal stability of the bio-based modified

GGM samples enhanced significantly compared to the unmodified GGM
specimen.
4. Conclusions
In summary, we developed a series of bio-based hemicellulose la­
texes, which can form films with high transparency, superior hydro­
phobicity and excellent stretchability. In the first stage of modification,
the A-GGM dispersions were prepared successfully by etherification
with DSal values ranging from 0.04 to 0.83. In the second stage, the semicontinuous emulsion polymerization technique was used to achieve free
radical polymerization between water soluble A-GGM and water insol­
uble n-butyl acrylate, thereby providing high molecular weight and long
hydrophobic chains. The resultant GGM-L coatings exhibited particle
sizes between 100 nm and 200 nm and were moderately stable, and
suitable for end-use applications such as packaging coatings on paper
and paperboard. Moreover, the optimal GGM-L film not only exhibited
high transmittance of 91.3% and haze of 0.43%, but also demonstrated
superior hydrophobicity arriving at 117.2◦ of water contact angle and
while maintaining excellent stretchability, with elongation at break of
31.2% and ultimate tensile stress of 30.9 MPa. Finally, the WVP of the
prepared GGM-L films was very close to the commercial HDPE and
showed a great increase in thermal stability compared to the pristine
GGM specimen. We anticipate that this newly-designed bio-based
hemicellulose-modified latex with bio-based content up to 99 wt% can
be industrially applied to replace the fossil-based polymer packaging
materials, which simultaneously boosts the exploitation and utilization
of hemicellulose wastes.

3.8. Thermal stability
Thermogravimetric analysis (TGA) was utilized to evaluate the
thermal stability of the unmodified GGM and modified GGM-based latex
films. As shown in Fig. 7a, an initial weight loss of less than 10% for all

the samples due to water evaporation is observed when the heating
temperature was below 220 ◦ C. All the samples started to degrade from
220 ◦ C, but the degradation stage and rate were quite different between
the unmodified GGM and modified GGM-L films, as indicated by the
derivative weight analysis in Fig. 7b. The pristine GGM film only shows
one large decomposition stage at around 300 ◦ C. This might be associ­
ated with disintegration of the macromolecular chains of GGM such as
the cleavage of the glycosidic bonds at 250 ◦ C, then followed by
degradation of the hexoses ranging from 250 ◦ C to 350 ◦ C (Mendes et al.,
2017).
However, the GGM-L films show two distinct stages of decomposi­
tion, one at around 220– 350 ◦ C and the other at 350– 450 ◦ C. The
temperature corresponding to maximum decomposition rate (T1max) at
around 300 ◦ C of GGM-based films increased with an increase of DS from
0.04 to 0.83 in the first stage, while in the second stage the temperature
of maximum decomposition rate (T2max) at 400 ◦ C remained constant.
This is might be due to the activation energy of formulated n-butyl
acrylate macromolecular main chains completely comprised of C–C
bonds was much higher than that of GGM backbones containing C-O-C
bonds (Hu, Chen, & Wang, 2004). Therefore, we can conclude that the
first decomposition stage of GGM-L films ranging from 220– 350 ◦ C can
be assigned to the degradation of GGM polymer backbone and the
cleavage of substituents, and the second decomposition stage ranging
from 350 ◦ C to 450 ◦ C is attributed to the degradation of n-butyl acrylate
polymer chains. Most of the degradation process was completed below
500 ◦ C. In Fig. 7a above 500 ◦ C, the unmodified GGM sample had a
higher char yield compared to GGM-based films. This was probably due
to differences in the salts content of the initial samples (Wang, Xiao, &
Lei, 2020). In general, through etherification and graft emulsion

CRediT authorship contribution statement
Conceptualization, experiment, data analysis and curation, writingoriginal draft, funding resources, Qiwen Yong; conceptualization,
experimental facility, review & editing, guidance and supervision,
funding resources, Martti Toivakka and Chunlin Xu；measurement,
data analysis, review, Jiayun Xu, Luyao Wang and Teija Tirri; supervi­
sion and modification, Hejun Gao, Yunwen Liao; All authors have read
and agreed to the published version of manuscript.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.

12


Q. Yong et al.

Carbohydrate Polymers 291 (2022) 119565

Acknowledgements

Mankar, A. R., Pandey, A., Modak, A., & Pant, K. K. (2021). Pretreatment of
lignocellulosic biomass: A review on recent advances. Bioresource Technology, 334,
Article 125235.
Markstedt, K., Xu, W., Liu, J., Xu, C., & Gatenholm, P. (2017). Synthesis of tunable
hydrogels based on O-acetyl-galactoglucomannans from spruce. Carbohydrate
Polymers, 157, 1349–1357.
Mendes, F. R. S., Bastos, M. S. R., Mendes, L. G., Silva, A. R. A., Sousa, F. D., MonteiroMoreira, A. C. O., & Moreira, R. A. (2017). Preparation and evaluation of
hemicellulose films and their blends. Food Hydrocolloids, 70, 181–190.
Mikkonen, K. S. (2020). Strategies for structuring diverse emulsion systems by using

wood lignocellulose-derived stabilizers. Green Chemistry, 22(4), 1019–1037.
Mikkonen, K. S., Heikkilă
a, M. I., Hel
en, H., Hyvă
onen, L., & Tenkanen, M. (2010). Spruce
galactoglucomannan films show promising barrier properties. Carbohydrate
Polymers, 79(4), 1107–1112.
Mikkonen, K. S., Heikkilă
a, M. I., Willfă
or, S. M., & Tenkanen, M. (2012). Films from
glyoxal-crosslinked spruce galactoglucomannans plasticized with sorbitol.
International Journal of Polymer Science, 2012, 1–8.
Mikkonen, K. S., Schmidt, J., Vesterinen, A.-H., & Tenkanen, M. (2013). Crosslinking
with ammonium zirconium carbonate improves the formation and properties of
spruce galactoglucomannan films. Journal of Materials Science, 48(12), 4205–4213.
Nechita, P., & Roman, M. (2020). Review on polysaccharides used in coatings for food
packaging papers. Coatings, 10(6), 566.
Niesbach, A., Fink, N., Lutze, P., & G´
orak, A. (2015). Design of reactive distillation
processes for the production of butyl acrylate: Impact of bio-based raw materials.
Chinese Journal of Chemical Engineering, 23(11), 1840–1850.
Peng, X., Du, F., & Zhong, L. (2019). Synthesis, characterization, and applications of
hemicelluloses based eco-friendly polymer composites. In Sustainable Polymer
Composites and Nanocomposites (pp. 1267–1322).
Qi, H., Chang, C., & Zhang, L. (2009). Properties and applications of biodegradable
transparent and photoluminescent cellulose films prepared via a green process.
Green Chemistry, 11(2), 177–184.
Rudawska, A., & Jacniacka, E. (2009). Analysis for determining surface free energy
uncertainty by the Owen-wendt method. International Journal of Adhesion and
Adhesives, 29(4), 451–457.

S´
arossy, Z., Tenkanen, M., Pitkă
anen, L., Bjerre, A.-B., & Plackett, D. (2013). Extraction
and chemical characterization of rye arabinoxylan and the effect of β-glucan on the
mechanical and barrier properties of cast arabinoxylan films. Food Hydrocolloids, 30
(1), 206–216.
Schatz, C., & Lecommandoux, S. (2010). Polysaccharide-containing block copolymers:
Synthesis, properties and applications of an emerging family of glycoconjugates.
Macromolecular Rapid Communications, 31(19), 1664–1684.
Song, Y., Xu, Y., Li, D., Chen, S., & Xu, F. (2021). Sustainable and superhydrophobic
lignocellulose-based transparent films with efficient light management and selfcleaning. ACS Applied Materials & Interfaces, 13(41), 49340–49347.
Sousa, S., et al. (2016). Xylan and xylan derivatives—Their performance in bio-based
films and effect of glycerol addition. Industrial Crops and Products, 94, 682–689.
/>Stark, N. M., & Matuana, L. M. (2021). Trends in sustainable biobased packaging
materials: A mini review. Materials Today Sustainability, 15, 9.
Tong, R., Chen, G., Tian, J., & He, M. (2020). Highly transparent, weakly hydrophilic and
biodegradable cellulose film for flexible electroluminescent devices. Carbohydrate
Polymers, 227, Article 115366.
Wang, R., Xiao, Y., & Lei, J. (2020). A solid-solid phase change material based on
dynamic ion cross-linking with reprocessability at room temperature. Chemical
Engineering Journal, 390, Article 124586.
Wei, D., Zeng, J., & Yong, Q. (2021). High-performance bio-based polyurethane
antismudge coatings using castor oil-based hyperbranched polyol as superior crosslinkers. ACS Applied Polymer Materials, 3(7), 3612–3622.
Wenzel, R. N. (2002). Resistance of solid surfaces to wetting by water. Industrial &
Engineering Chemistry, 28(8), 988–994.
Yang, H., Yi, N., Zhao, S., Xiang, Z., Qaseem, M. F., Zheng, B., & Wu, A. M. (2021).
Characterization of hemicellulose in cassava (Manihot esculenta Crantz) stem during
xylogenesis. Carbohydrate Polymers, 264, Article 118038.
Yi, Y., Xu, W., Wang, H. X., Huang, F., & Wang, L. M. (2020). Natural polysaccharides
experience physiochemical and functional changes during preparation: A review.

Carbohydrate Polymers, 234, Article 115896.
Yong, Q., & Liang, C. (2019). Synthesis of an aqueous self-matting acrylic resin with low
gloss and high transparency via controlling surface morphology. Polymers (Basel), 11
(2), 322.
Yong, Q., Nian, F., Liao, B., Guo, Y., Huang, L., Wang, L., & Pang, H. (2016). Synthesis
and surface analysis of self-matt coating based on waterborne polyurethane resin
and study on the matt mechanism. Polymer Bulletin, 74(4), 1061–1076.
Zasadowski, D., Yang, J., Edlund, H., & Norgren, M. (2014). Antisolvent precipitation of
water-soluble hemicelluloses from TMP process water. Carbohydrate Polymers, 113,
411–419.
Zhao, H., Mikkonen, K. S., Kilpelainen, P. O., & Lehtonen, M. I. (2020). Spruce
galactoglucomannan-stabilized emulsions enhance bioaccessibility of bioactive
compounds. Foods, 9(5), 672.
Zhu, H., Parvinian, S., Preston, C., Vaaland, O., Ruan, Z., & Hu, L. (2013). Transparent
nanopaper with tailored optical properties. Nanoscale, 5(9), 3787–3792.
Zoldners, J., & Kiseleva, T. (2013). Modification of hemicelluloses with polycarboxylic
acids. Holzforschung, 67(5), 567–571.

This work was financially supported by China Scholarship Council
(202008510014), research fund of China West Normal University, China
(Grant No. 19B027), and Business Finland (2320/31/2021). All re­
searchers in Laboratory of Natural Materials Technology are acknowl­
edged for research assistance to support the current work. This work is
also part of activities within the Johan Gadolin Process Chemistry
Centre (PCC) at ÅAU.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.carbpol.2022.119565.
References
Alekhina, M., Mikkonen, K. S., Alen, R., Tenkanen, M., & Sixta, H. (2014).

Carboxymethylation of alkali extracted xylan for preparation of bio-based packaging
films. Carbohydrate Polymers, 100, 89–96.
Battista, F., Zuliani, L., Rizzioli, F., Fusco, S., & Bolzonella, D. (2021). Biodiesel, biogas
and fermentable sugars production from spent coffee grounds: A cascade biorefinery
approach. Bioresource Technology, 342, Article 125952.
Chadni, M., Grimi, N., Bals, O., Ziegler-Devin, I., & Brosse, N. (2019). Steam explosion
process for the selective extraction of hemicelluloses polymers from spruce sawdust.
Industrial Crops and Products, 141, Article 111757.
Chio, C., Sain, M., & Qin, W. (2019). Lignin utilization: A review of lignin
depolymerization from various aspects. Renewable and Sustainable Energy Reviews,
107, 232–249.
Cunha, A. G., & Gandini, A. (2010). Turning polysaccharides into hydrophobic materials:
A critical review. Part 2. Hemicelluloses, chitin/chitosan, starch, pectin and
alginates. Cellulose, 17(6), 1045–1065.
Farhat, W., Venditti, R., Ayoub, A., Prochazka, F., Fern´
andez-de-Alba, C., Mignard, N.,
Taha, M., & Becquart, F. (2018). Towards thermoplastic hemicellulose: Chemistry
and characteristics of poly-(ε-caprolactone) grafting onto hemicellulose backbones.
Materials & Design, 153, 298307.
Garcớa, V., Pă
akkilă
a, J., Ojamo, H., Muurinen, E., & Keiski, R. L. (2011). Challenges in
biobutanol production: How to improve the efficiency? Renewable and Sustainable
Energy Reviews, 15(2), 964–980.
Geng, W., Venditti, R. A., Pawlak, J. J., Chang, H.-M., Pal, L., & Ford, E. (2020).
Carboxymethylation of hemicellulose isolated from poplar (Populus grandidentata)
and its potential in water-soluble oxygen barrier films. Cellulose, 27(6), 3359–3377.
Hansen, N. M., & Plackett, D. (2008). Sustainable films and coatings from hemicelluloses:
A review. Biomacromolecules, 9(6), 1493–1505.
Hardelin, L., Bernin, D., Borjesson, M., Strom, A., & Larsson, A. (2020). Altered thermal

and mechanical properties of spruce galactoglucomannan films modified with an
etherification reaction. Biomacromolecules, 21(5), 1832–1840.
Hu, Y. H., Chen, C. Y., & Wang, C. C. (2004). Thermal degradation kinetics of poly(nbutyl acrylate) initiated by lactams and thiols. Polymer Degradation and Stability, 84
(3), 505–514.
Huang, J., Zhong, Y., Zhang, L., & Cai, J. (2017). Extremely strong and transparent chitin
films: A high-efficiency, energy-saving, and “Green” route using an aqueous KOH/
Urea solution. Advanced Functional Materials, 27(26), 1701100.
Khwaldia, K., Arab-Tehrany, E., & Desobry, S. (2010). Biopolymer coatings on paper
packaging materials. Comprehensive Reviews in Food Science and Food Safety, 9(1),
8291.
Kisonen, V., Xu, C., Bollstră
om, R., Hartman, J., Rautkoski, H., Nurmi, M., Hemming, J.,
Eklund, P., & Willfă
or, S. (2014). O-acetyl galactoglucomannan esters for barrier
coatings. Cellulose, 21(6), 4497–4509.
Laine, C., Harlin, A., Hartman, J., Hyvă
arinen, S., Kammiovirta, K., Krogerus, B., & Vă
ahă
aNissi, M. (2013). Hydroxyalkylated xylans Their synthesis and application in
coatings for packaging and paper. Industrial Crops and Products, 44, 692704.
Lehtonen, M., Teră
aslahti, S., Xu, C., Yadav, M. P., Lampi, A.-M., & Mikkonen, K. S.
(2016). Spruce galactoglucomannans inhibit lipid oxidation in rapeseed oil-in-water
emulsions. Food Hydrocolloids, 58, 255–266.
Leppanen, A. S., Xu, C., Parikka, K., Eklund, P., Sjoholm, R., Brumer, H., & Willfor, S.
(2014). Targeted allylation and propargylation of galactose-containing
polysaccharides in water. Carbohydrate Polymers, 100, 46–54.
Liu, R., Du, J., Zhang, Z., Li, H., Lu, J., Cheng, Y., & Wang, H. (2019). Preparation of
polyacrylic acid-grafted-acryloyl/hemicellulose (PAA-g-AH) hybrid films with high
oxygen barrier performance. Carbohydrate Polymers, 205, 83–88.

Liu, X., Renard, C., Bureau, S., & Le Bourvellec, C. (2021). Revisiting the contribution of
ATR-FTIR spectroscopy to characterize plant cell wall polysaccharides. Carbohydrate
Polymers, 262, Article 117935.
Maleki, L., Edlund, U., & Albertsson, A. C. (2017). Synthesis of full interpenetrating
hemicellulose hydrogel networks. Carbohydrate Polymers, 170, 254–263.

13