The use of N,N′-diallylaldardiamides as cross-linkers in xylan derivatives-based hydrogels

Article abstract of DOI:10.1016/j.carres.2011.09.028

N,N′-Diallylaldardiamides (DA) were synthesized from galactaric, xylaric, and arabinaric acids, and used as cross-linkers together with xylan (X) derivatives to create new bio-based hydrogels. Birch pulp extracted xylan was derivatized to different degrees of substitution of 1-allyloxy-2-hydroxy-propyl (A) groups combined with 1-butyloxy-2-hydroxy-propyl (B) and/or hydroxypropyl (HP) groups. The hydrogels were prepared in water solution by UV induced free-radical cross-linking polymerization of derivatized xylan polymers without DA cross-linker (xylan derivative hydrogel) or in the presence of 1 or 5 wt % of DA cross-linker (DA hydrogel). Commercially available cross-linker (+)-N,N′-diallyltartardiamide (DAT) was also used. The degree of substitution (DS) of A, B, and HP groups in xylan derivatives was analyzed according to ¹H NMR spectra. The DS values for the cross-linkable A groups of the derivatized xylans were 0.4 (HPX-A), 0.2 (HPX-BA), and 0.4 (X-BA). The hydrogels were examined with FT-IR and elemental analysis which proved the cross-linking successful. Water absorption of the hydrogels was examined in deionized water. Swelling degrees up to 350% were observed. The swollen morphology of the hydrogels was assessed by scanning electron microscopy (SEM). The presence of cross-linkers in DA hydrogels had only a small impact on the water absorbency when compared to xylan derivative hydrogels but a more uniform pore structure was achieved.

Full text of DOI:10.1016/j.carres.2011.09.028

Carbohydrate Research 346 (2011) 2736–2745
Contents lists available at SciVerse ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
The use of N,N0-diallylaldardiamides as cross-linkers in xylan
derivatives-based hydrogels
⇑
Helinä Pohjanlehto, Harri Setälä , Kari Kammiovirta, Ali Harlin
Technical Research Centre of Finland (VTT), Biologinkuja 7, FI-02044 Espoo, Finland
a r t i c l e i n f o
a b s t r a c t
N,N0-Diallylaldardiamides (DA) were synthesized from galactaric, xylaric, and arabinaric acids, and used
as cross-linkers together with xylan (X) derivatives to create new bio-based hydrogels. Birch pulp
extracted xylan was derivatized to different degrees of substitution of 1-allyloxy-2-hydroxy-propyl (A)
groups combined with 1-butyloxy-2-hydroxy-propyl (B) and/or hydroxypropyl (HP) groups. The hydro-
gels were prepared in water solution by UV induced free-radical cross-linking polymerization of deriva-
tized xylan polymers without DA cross-linker (xylan derivative hydrogel) or in the presence of 1 or 5 wt %
of DA cross-linker (DA hydrogel). Commercially available cross-linker (+)-N,N0-diallyltartardiamide (DAT)
was also used. The degree of substitution (DS) of A, B, and HP groups in xylan derivatives was analyzed
according to 1H NMR spectra. The DS values for the cross-linkable A groups of the derivatized xylans were
0.4 (HPX-A), 0.2 (HPX-BA), and 0.4 (X-BA). The hydrogels were examined with FT-IR and elemental anal-
ysis which proved the cross-linking successful. Water absorption of the hydrogels was examined in
deionized water. Swelling degrees up to 350% were observed. The swollen morphology of the hydrogels
was assessed by scanning electron microscopy (SEM). The presence of cross-linkers in DA hydrogels had
only a small impact on the water absorbency when compared to xylan derivative hydrogels but a more
uniform pore structure was achieved.
Article history:
Received 16 June 2011
Received in revised form 21 September 2011
Accepted 23 September 2011
Available online 2 October 2011
Keywords:
Hydrogel
N,N0-Diallylaldardiamides
Xylan
Renewable resources
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
they are normally disposed of as organic waste from the forest
industry sidestreams. However, recent research has begun to find
Hydrogels are chemically or physically cross-linked networks
that are water-insoluble but capable of absorbing large amounts
of water.1 They can be made of synthetic or natural starting mate-
rials but commercial hydrogels have traditionally been prepared
mainly from toxic acrylates and acrylamides.2 Hydrogels based on
naturally occurring products are of interest not only for their
renewable character and nontoxic nature but because they may of-
fer biocompatibility and biodegradability. Hydrogels possess a de-
gree of flexibility due to their significant water content and they
are potential material candidates, for example, in tissue engineer-
ing,3 controlled drug release,4,5 agriculture,6,7 and hygiene prod-
ucts.8 Especially beneficial for future applications is the obvious
biodegradability of polysaccharide and aldaric acid based materials.
Carbohydrates are a class of natural products that are widely
available in nature. They have a wide range of functionalities and
are very hydrophilic, which makes them good candidates for
hydrogel preparation. Hemicelluloses, such as xylan, have hydroxyl
groups in each repeating unit that can be chemically derivatized to
new reacting groups. When compared with cellulose and starch,
hemicelluloses have been somewhat neglected in research and
new applications for hemicelluloses and examples of hydrogels
prepared from modified hemicelluloses can be found.9–13
Hemicelluloses can also be hydrolyzed to monosaccharides to
obtain, for example, xylose, arabinose, galactose, and mannose that
can be oxidized to aldaric acids. Aldaric acids are typically pre-
pared by simple chemical oxidations of sugars, but can also be pro-
duced biotechnically.14 These diacids can be used as starting
materials, for example, in the preparation of diallyl functionalized
cross-linkers or other derivatives.15 The aim of this work was to
synthesize analogues of the commercially available cross-linker
(+)-N,N0-diallyltartardiamide (DAT) and further use them as novel
cross-linkers in the preparation of bio-based hydrogels. Some syn-
theses of N,N0-diallylaldardiamides (DA) have been described in
the literature but none starting with bio-based starting materials
or used in cross-linking reactions. Anker16 has reported the prepa-
ration of polyacrylamide gels using DAT as a cross-linker instead of
the normally used methylenebisacrylamide (MBA). Two references
of N,N0-diallylgalactardiamide (DAG)17,18 and one of N,N0-diallylxy-
lardiamide (DAX)17 were found. In contrast, no references for N,N0-
diallylarabinardiamide (DAA) were found in the literature so far.
We report here novel xylan based hydrogels prepared in water
solution by cross-linking xylan derived polymers with and with-
out N,N0-diallylaldardiamide cross-linkers. Three different xylan
⇑
Corresponding author. Tel.: +358 50 520 7979; fax: +358 20 722 7026.
E-mail address: harri.setala@vtt.fi (H. Setälä).
0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.09.028

H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
2737
derivatives were prepared: HPX-A, HPX-BA, and X-BA. HPX-A and
HPX-BA were prepared from xylan by first attaching hydroxypro-
pyl (HP), and subsequently 1-allyloxy-2-hydroxy-propyl (A) or
both 1-allyloxy-2-hydroxy-propyl (A) and 1-butyloxy-2-hydro-
xy-propyl (B) groups, respectively. X-BA was prepared from xylan
by attaching directly A and B groups to the xylan backbone. 1-
can be derivatized to different DS-values with varying derivatizing
agents. Hydroxypropylation increases the hydrophilicity of xylan
thus improving the solubility of xylan to water.25 Hydroxypropyla-
tion of xylan was performed according to the literature.25 Hydroxy-
propylated or unmodified xylan was reacted in alkaline conditions
with allyl glycidyl ether and/or butyl glycidyl ether to obtain three
different xylan derivatives (HPX-BA, HPX-A, and X-BA) with differ-
ent structures and DS-values of allyl groups and other substituents
(Figs. 2–4). 1-Allyloxy-2-hydroxy-propyl (A) group is a cross-link-
able functionality. In addition, the A and 1-butyloxy-2-hydroxy-
propyl (B) groups form a similar structure compared to free HP
groups with a secondary hydroxyl group in carbon-2 thus also
improving the water solubility. X-BA prepared from unmodified
xylan without hydroxypropylation had a similar water solubility
compared with HPX-BA and HPX-A. All the xylan derivatives were
soluble into water up to 15–20 wt %. The effect of the non-reacting
B group similar in size to A group in hydrogels was studied to
investigate the reactivity differences and water absorption capaci-
ties. The DS-values of the xylan derivatives were determined by the
integration of 1H NMR spectra.26 In addition, DS-values of allyl
groups were verified using a bromination method. DS-values are
presented in Table 2.
Allyloxy-2-hydroxy-propyl (A) group is
a cross-linkable allyl
functionality which was used as the cross-linking site in this re-
search. Polysaccharides have previously been derivatized with
similar groups, for example, Huijbrechts et al.19 have modified
starch with allyl glycidyl ether and compared its physicochemical
properties compared to native starch. Shen et al.20 have modified
carboxymethyl cellulose to obtain an allyl functionalized deriva-
tive, N-allylcarbamoylmethyl cellulose, for hydrogel preparation.
We chose aldaric acid based cross-linkers not only because they
are bio-based but because this permits cross-linking reaction in
water. Derivatized xylan was cross-linked without DA cross-lin-
ker as well as with four different types of cross-linkers (DAT,
DAX, DAA, and DAG). The morphological and swelling properties
of the hydrogels were determined.
2. Results and discussion
The hydrogels were prepared by cross-linking the xylan deriva-
tives without the DA cross-linker or by inserting the DA cross-lin-
ker between the xylan polymer chains. When no additional DA
cross-linker is present the cross-linking reaction takes place be-
tween the A groups of the modified xylan polymers. The cross-link-
ing was done in a 10% water solution of the xylan polymer. The
derivatized xylan polymers dissolved well in water and the ob-
tained solutions were slightly cloudy. The cross-linkers were dis-
solved in water prior to mixing with the polymer solution.
Potassium persulfate was used as the photoinitiator and it was dis-
solved in a small amount of water prior to mixing with the solution
of xylan polymer and the cross-linker. Samples were put on Petri
dishes and exposed to UV light in an UV oven. The samples were
irradiated in 30 s periods and let to cool down in between. The gels
were irradiated for 3–4 min (Fig. 5).
2.1. Synthesis of xylan hydrogels
Three different cross-linkers, DAX, DAA, and DAG, were synthe-
sized from aldaric acids. Commercially available cross-linker (+)-
N,N0-diallyltartardiamide (DAT) was also used. The reaction route
for the synthesis of the cross-linkers is shown schematically in
Scheme 1. Xylaric (1a) and arabinaric (1b) acids were synthesized
according to the literature21 from xylose and arabinose, respec-
tively. Galactaric acid (1c) was commercially available. All the dia-
cids were esterified22 with methanol to the corresponding diesters
(2a, 2b, and 2c) and subsequently reacted with allylamine16 in dry
tetrahydrofuran to obtain white crystalline cross-linkers DAX (3a),
DAA (3b), and DAG (3c). The yield of 3c (59%) was better than the
yields of 3a (37%) and 3b (32%). This can be due to the fact that the
starting materials for the reaction with allylamine, 2a and 2b, were
syrupy products with some solvent or other small molecule impu-
rities and were used in the synthesis as such with no additional
purification steps.
2.2. Characterization
A representative FT-IR spectra of a derivatized xylan hydrogel
prepared with no additional DA cross-linker (red spectrum) and
the DA cross-linked hydrogel (black spectrum) are shown in Figure
6. The spectra were quite similar since the only new functional
group introduced by the cross-linking reaction with DA cross-link-
ers was the amide group. A differing amide peak was seen at
1538 cmꢀ1 (black spectrum).
DA hydrogels were analyzed for their nitrogen content to inves-
tigate the reactivity and incorporation of the DA cross-linkers into
the hydrogel network. The DA cross-linkers contain an amide
group allowing elemental analysis as a means to prove the incor-
poration of the cross-linkers in between the derivatized xylan
All cross-linkers were verified by 1H NMR. A representative 1H
NMR spectrum of a DA cross-linker (DAG) is shown in Figure 1.
The cross-linkers were soluble in DMSO and water. No peaks are
seen from methyl ester which indicates a successful reaction with
allylamine. The amide proton gives a peak at 7.73 ppm and the
double bond gives peaks at 5.84 and 5.14 ppm. The methylene pro-
tons next to the amide group can be seen at 3.78 ppm.
Non-dried (5–11 wt % of xylan, containing 0.9% NaOH) or dried
xylans extracted, for example, from bleached birch pulp23,24 were
used as starting materials in the preparation of hydrogels. Xylan
polymer repeating unit has two secondary hydroxyl groups, which
O
O
O
O
O
O
MeOH
H2N
HO
OH
MeO
OMe
N
H
N
H
n
H2SO4
dry THF
24h, 70 oC
n
n
24h, 60 oC
OH
OH
OH
1a: (n=3) Xylaric acid
2a: (n=3) Dimethyl xylarate
3a: (n=3) DAX
3b: (n=3) DAA
3c: (n=4) DAG
3d: (n=2) DAT
1b: (n=3) Arabinaric acid
1c: (n=4) Galactaric acid
2b: (n=3) Dimethyl arabinarate
2c: (n=4) Dimethyl galactarate
Scheme 1. A schematic representation of the preparation of DA cross-linkers.

2738
H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
OH
OH
O
H
N
3
1
5
4
N
H
6
2
O
OH
OH
Figure 1. 1H NMR spectrum of DAG recorded in DMSO-d6.
chains. Varying amounts of nitrogen were found. Differences in
nitrogen contents may be due to reactivity differences of the DA
cross-linkers. The elemental analysis for the samples cross-linked
with 5 wt % of cross-linker gave increased values of nitrogen, a
clear indication of a successful cross-linking reaction. The incorpo-
ration efficiency for all of the DA cross-linkers was calculated from
the elemental analysis results as shown in Table 1. The incorpora-
tion efficiencies of DAA, DAX, and DAG cross-linkers are compara-
ble to the results of the commercially available cross-linker DAT. X-
BA hydrogels seem to have the lowest and HPX-A hydrogels the
highest incorporation efficiency of DA cross-linkers. The higher
incorporation efficiency of DA cross-linkers into HPX-A hydrogels
can be due to the fact that there are no interfering butyl groups
during the cross-linking reaction thus allowing the allylic double
bonds to react more freely and also due to lower cross-linking effi-
ciency of allylic double bonds among themselves (see Table 2). The
results from the elemental analysis for the samples with 1 wt % of
cross-linker were below detection limit and therefore could not be
reliably interpreted.
substituents in a xylan derivative.26,27 DS-values of other substit-
uents (B and HP) were calculated from 1H NMR measurements.
DS-values using the bromination method were slightly lower than
DS-values calculated from 1H NMR measurements as observed
previously by Lepistö et al.28 The reason may be that the bromin-
ation reaction is not quantitative and/or 1H NMR measurement
gives too high DS-values. However, the two methods gave compa-
rable results.
SEM was used to characterize the morphology and structural
differences between the derivatized xylan hydrogels and the DA
hydrogels. The hydrogel samples were immersed in liquid nitrogen
and freeze-dried before the SEM analysis. Typical images of the
derivatized xylan hydrogels and the DA hydrogels are presented
in Figures 8 and 9, respectively. The hydrogels with DA cross-link-
ers clearly had a more uniform and even pore structure as com-
pared to the derivatized xylan hydrogels with no additional
cross-linkers. The same pattern in the morphology could be ob-
served for all the hydrogels.
The incorporation of the DA cross-linkers was also apparent
from the transparency of the hydrogels: the gel with 5 wt % of
cross-linker (left) was whiter and almost opaque whereas the gel
with 1 wt % of cross-linker (right) was semi-opaque (Fig. 7). The
gels with no additional cross-linker were almost completely trans-
parent (data not shown).
2.3. Swelling measurements
Swelling measurements were performed for all synthesized
hydrogels by placing the dried gels in an excess of deionized water
at room temperature. Figure 10 shows the differences in absorben-
cies between the derivatized xylan hydrogels (X-BA, HPX-BA, and
HPX-A) with no additional cross-linkers. It was expected that the
gels made with xylan derived polymer with the lowest DS of
cross-linkable allyl groups (0.2 for HPX-BA) would have the loosest
network of cross-links and thus the highest water absorbency.
However, water absorbency was highest for HPX-A and the lowest
for X-BA both of which have a DS of allyl groups 0.4 based on the
NMR measurements. According to cross-linking efficiencies, X-BA
and HPX-BA form a more dense network and thus absorb less
water than HPX-A, a similar phenomenon observed by Voepel
et al.29 when preparing O-acetyl galactoglucomannan hydrogels.
The lower water absorbency can also be due to the presence of bu-
tyl groups in HPX-BA and X-BA polymers that makes them more
hydrophobic. This is in accordance with previous work done with
xylans where the water absorbency decreased after bringing
hydrophobic groups to the xylan backbone.13 The hydroxypropyl
groups in the xylan backbone of HPX-A and HPX-BA also seemed
to have a positive influence on the water absorption due to longer
side chains enabling a more open structure.
The cross-linking efficiency of allyl functionalities in the xylan
derivatives without DA cross-linkers was demonstrated using a
bromination method of allylic double bonds with bromine.27 Table
2 shows the bromine content of the analyzed xylan derivatives. As
xylan does not contain double bonds, it was used as the reference
sample and accordingly no bromine was found. When comparing
the starting material HPX-A to the cross-linked HPX-A, a decrease
in the amount of bromine from 13.7 wt % to 11.3 wt % was seen,
which indicates that double bonds have reacted. The cross-linking
efficiency was calculated to be 18%. Similarly, the bromine content
was observed to decrease from 6.5 wt % of HPX-BA to 4.4 wt% of the
cross-linked HPX-BA, and from 10.9 wt % of X-BA to 5.4 wt % of the
cross-linked X-BA. The cross-linking efficiencies were 32% and 50%,
respectively. Allylic double bonds of X-BA seem to cross-link best
among themselves leading to a less efficient incorporation of the
DA cross-linkers into the hydrogel network. In conclusion, when
the cross-linking efficiency of the allyl functionalities among them-
selves is lower, the incorporation of the DA cross-linkers is higher
(see Tables 1 and 2).
Hydrogels with DA cross-linkers showed the same swelling
trend than the hydrogels with no additional cross-linker. The
amount (1 or 5 wt %) and type of DA cross-linker used had some
DS-values of allylic double bonds were calculated using the
bromination method taking into account the amounts of other

H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
2739
A
24
23
HP
O
22
OH
21
20
27
26
25
HP
19
OH
O
O
17
16
18
4
O
5
O
8
O
O
O
1
O
2
3
OH
7
6
O
10
9
12
11
14
O
HP
OH
13
15
B
Figure 2. 1H NMR spectrum of HPX-BA.
influence on the water absorbency (Fig. 11). A general trend for
hydrogels prepared from X-BA was observed: the gels prepared
with 5 wt % of any DA cross-linker absorbed less water than those
prepared with 1 wt % cross-linker. This could be due to the higher
density of the network in the gels prepared with 5 wt % cross-lin-
ker and could be expected also for the other hydrogels.30 However,
the hydrogels prepared with HPX-A showed no significant differ-
ences in the absorbencies between gels prepared with 1 or
5 wt % of any DA cross-linker. Only the HPX-A hydrogel cross-
linked with DAX showed a similar swelling behavior than the X-
BA hydrogels. HPX-BA hydrogels had varying swelling properties.
The gels cross-linked with 1 or 5 wt % of DAG or DAX had no signif-
icant differences in swelling properties, whereas the gels cross-
linked with 5 wt % of DAT absorbed more water than the gels
cross-linked with 1 wt % of DAT. The gels cross-linked with DAA
showed the opposite effect.
3. Experimental
3.1. Materials
The following reagents were used as received: (+)-N,N0-diallyltar-
tardiamide (99+%, Aldrich), -(+)-xylose (P99%, Sigma–Aldrich),
-(+)-arabinose (99%, Sigma), galactaric acid (mucic acid, 97%, Al-
D
L
drich), allylamine (P98.0%, Fluka), tetrahydrofuran (99.9%, Aldrich),
methanol (HPLC grade, Rathburn), ethanol (Ba, Altia), sulfuric acid
(95–97%, Fluka), nitric acid (P65%, Fluka), 2-propanol (HPLC grade,
Rathburn), allyl glycidyl ether (P99%, Sigma–Aldrich), butyl glyc-
idyl ether (95% Aldrich), propylene oxide (99%, Sigma–Aldrich),
NaOH (30% w/w, Reagecon), HCl (37%, Sigma–Aldrich), acetone
(technical grade, YA-kemia), t-BuOH (>99.0%, Fluka). The following
starting materials were synthesized according to literature:
hydroxypropylated xylan,25 dimethylgalactarate, dimethylxylarate

2740
H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
O
OH
OH
O
O
O
O
O
O
O
O
OH
O
O
OH
Figure 3. 1H NMR spectrum of HPX-A.
and dimethylarabinarate,22 and xylaric and arabinaric acid.21 The
cross-linkers were prepared according to Anker.16 All the cross-link-
ers were soluble to water but the dissolution of DAG required some
heating. Never-dried (5–11 wt % of xylan, containing 0.9% NaOH) or
dried xylans extracted, for example, from (bleached) birch pulp were
used as starting materials in the preparation of hydrogels.
MicroMass Quattro II Spectrometer. An F300S UV system (Fusion
UV Systems, Inc.) was used for the irradiation of the hydrogel sam-
ples. The UV system consists of a P300MT power supply and an
I300MB irradiator which emits UV light in a range from 200 to
400 nm. The structures of the swollen gels were characterized by
SEM (LEO DSM 982 Gemini FEG-SEM).
3.2. Methods
3.3. Synthesis of the monomers
1H and 13C NMR spectra were recorded on a Varian Mercury VX
(300 MHz) spectrometer or on a Bruker Avance III (500 MHz) spec-
trometer in DMSO or D2O. Melting points (mp) were determined in
open capillaries using SANYO Gallenkamp mp apparatus. Elemen-
tal analyses were performed on Elementar Analysensysteme
GmbH Variomax CHN analyzer. Bromine samples were irradiated
in the TRIGA MARK II reactor followed by analysis with an auto-
matic HPGe-detector attached to an automatic gamma spectrome-
ter. The accuracy of the bromination method is 10%. IR spectra
were recorded on a Bruker Equinox 55 FTIR spectrometer. Mass
spectra were recorded using direct infusion ESI technique on
3.3.1. N,N0-Diallylarabinardiamide (3b)
L
-(+)-Arabinose was oxidized21 to
L-(+)-arabinaric acid (1b) with
concentrated HNO3 and subsequently esterified22 with methanol
according to Kiely et al. The resulting syrupy dimethylarabinarate
2b (21.9 g, 0.105 mol) was dissolved under argon in dry tetrahy-
drofuran (210 mL) in 500 mL glass flask. Allylamine (24.5 mL,
0.326 mol) was added and the temperature was raised to 70 °C.
After 22 h the solution was cooled and the crystals were filtered
and washed with THF/10% ethanol-solution (110 mL). Light yellow
crystals were obtained, yield 8.6 g (32%): mp 180–185 °C; IR:
3280 (amide I), 3082 (CH@), 1638 (CO), 1553 (amide II), 1045
m

H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
2741
O
HO
OH
HO
O
O
O
O
O
O
OH
O
Figure 4. 1H NMR spectrum of X-BA.
(OH) cmꢀ1 1H NMR (Me2SO-d6, 300 MHz): d 8.10 (t, 1H, NH,
.
J = 11.4 Hz, J = 6.0 Hz), 7.76 (t, 1H, NH, J = 6.1 Hz, J = 12.5 Hz), 5.82
(m, 2H, allyl =CH, J = 1.8 Hz), 5.62 (d, 2H, OH-2/4, J = 6.2 Hz), 5.12
(m, 4H, allyl CH2@, J = 1.8 Hz), 4.77 (d, 1H, OH-3, J = 6.3 Hz), 4.15
(dd, 2H, H-3, J = 1.5 Hz, J = 6.1 Hz), 3.95 (m, 2H, H-2/4, J = 1.5 Hz,
J = 6.3 Hz), 3.77 (t, 4H, –CH2–NH, J = 5.5 Hz); 13C NMR (Me2SO-d6,
75 MHz): 174.2, 173.3 (CONH), 136.3, 136.0 (–CH@CH2), 116.0,
115.8 (–CH@CH2), 73.9 (C-3), 72.4, 72.3 (C-2 ja C-4), 41.6 (CH2–
NH); MS (ESI): [M+H]+ found 259; requires 259.282. Anal. Calcd
for C11H18O5N2: C, 51.15; H, 7.03; N, 10.85. Found: C, 40.42; H,
5.69; N, 8.08.
3.3.2. N,N0-Diallylxylardiamide (3a)
The starting material dimethylxylarate (2a) was prepared from
Figure 5. Hydrogel prepared from HPX-A with 5 wt % of DAA cross-linker.
D
-(+)-xylose as described in Section 3.3.1. Compound 2a (15.0 g,
166–170 °C).17 IR:
(amide II), 1125 (OH) cmꢀ1
7.83 (t, 2H, NH, J = 5.9 Hz, J = 11.9), 5.84 (m, 2H, allyl @CH,
J = 1.8 Hz), 5.50 (d, 2H, OH-2/4, J = 5.7 Hz), 5.13 (m, 4H, allyl
CH2@, J = 1.8 Hz), 4.81 (d, 1H, OH-3, J = 7.3 Hz), 4.11 (dd, 2H, H-2/
4, J = 5.7 Hz, J = 4.1 Hz), 3.95 (m, 1H, H-3, J = 7.3 Hz, J = 4.1 Hz),
m
3417 (amide I), 3080 (CH@), 1638 (CO), 1546
1H NMR (Me2SO-d6, 300 MHz): d
71.9 mmol) was dissolved under argon in dry tetrahydrofuran
(150 mL) in 250 mL glass flask. Allylamine (16.2 mL, 215.6 mmol)
was added and the temperature was raised to 70 °C. After 24 h
the solution was cooled and the crystals were filtered and washed
with THF/10% ethanol-solution (80 mL). Light yellow crystals were
recrystallized from ethanol, yield 7.0 g (37%): mp 181–185 °C (lit.:
.

2742
H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
Figure 6. IR spectrum of HPX-BA (red spectrum) and the spectrum of HPX-BA cross-linked with 5 wt % of DAT (black spectrum).
Table 1
The incorporation efficiency (E) of DA cross-linkers calculated from elemental
analysis of hydrogels cross-linked with 5 wt % of cross-linker
HPX-A (wt %) E (%) HPX-BA (wt %) E (%) X-BA (wt %) E (%)
DAT
3.86
77
74
71
53
3.79
3.34
2.96
2.22
76
67
59
44
2.07
1.67
1.56
1.48
41
33
31
30
DAG 3.72
DAA 3.53
DAX 2.63
Table 2
Crosslinking efficiency of allylic double bonds calculated from the amounts of
bromine found with instrumental neutron activation method
Sample
Br-content
(wt %)
DSA
(Br)
DSA
DSB
DSHP
(1H NMR)
(1H NMR)
(1H NMR)
Xylan
HPX-A
HPX-A
cross-linked
HPX-BA
HPX-BA
cross-linked
X-BA
X-BA
0.04
13.7
11.3
0
0.20
0
0.36
0
0
0.85
0.63
0
Figure 8. SEM images of X-BA hydrogel without DA cross-linker. The scale is
20 m.
6.5
4.4
0.10
0.20
0.17
0.38
0.34
0.75
l
10.9
5.4
cross-linked
Degrees of substitution determined by the bromination and 1H NMR methods are
also presented.
Figure 7. The opaqueness of HPX-BA hydrogels with 5 wt % (left) and 1 wt % (right)
of DA cross-linker.
Figure 9. SEM image of X-BA hydrogel with 1 wt % of DAX cross-linker. The scale is
20 m.
l

H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
2743
Figure 10. Swelling curves of X-BA, HPX-BA, and HPX-A without DA cross-linkers.
3.76 (m, 4H, –CH2–NH, J = 5.9 Hz); 13C NMR (Me2SO-d6, 75 MHz):
173.3 (CONH), 136.2 (–CH@CH2), 115.9 (–CH@CH2), 73.5 (C-2 ja
C-4), 72.9 (C3), 41.6 (CH2–NH); MS (ESI): [M+H]+ found 259; re-
quires 259.282. Anal. Calcd for C11H18O5N2: C, 51.15; H, 7.03; N,
10.85. Found: C, 46.28; H, 6.38; N, 9.52.
3.3.3. N,N0-Diallylgalactardiamide (3c)
Galactaric acid (1c) was esterified with methanol according to
Kiely et al.22 Dimethylgalactarate (2c) (26.5 g, 111.3 mmol) was
dispersed in dry tetrahydrofuran (400 mL) under argon in 1 L glass
reactor. Allylamine (25.1 mL, 334 mmol) was added and the tem-
perature was raised to 70 °C. After 48 h the solution was cooled
and the crystals filtered and washed with THF/10% ethanol-solu-
tion (300 mL). Recrystallization from water gave pure white crys-
tals, yield 19.1 g (59%): mp 215–220 °C (lit. 207–210 °C);17 IR:
3290 (amide I), 3084 (CH@), 1634 (CO), 1537 (amide II), 1047
(OH) cmꢀ1 1H NMR (Me2SO-d6, 300 MHz): d 7.73 (t, 2H, –NH,
m
.
J = 12.1 Hz, J = 6.0 Hz), 5.84 (m, 2H, allyl @CH, J = 5.1 Hz,
J = 10.2 Hz,) 5.28 (d, 2H, OH-2/5, J = 7.1 Hz), 5.14 (m, 4H, allyl
CH2@, J = 5.1, J = 10.3), 4.45 (dd, 2H, OH-3/4, J = 2.5, J = 8.6), 4.2
(d, 2H, C-2/5, J = 7.1), 3.84 (dd, 2H, C-3/4, J = 2.1, J = 8.6), 3.78 (t,
4H, –CH2–NH, J = 5.5, J = 11.0 Hz); 13C NMR (Me2SO-d6, 75 MHz):
174.2 (CONH), 136.3 (–CH@CH2), 115.8 (–CH@CH2), 71.7 (C-2/3/
4/5), 41.6 (CH2 –NH); MS (ESI): [M+H]+ found 289; requires
289.308. Anal. Calcd for C12H20O6N2: C, 49.99; H, 6.99; N, 9.72.
Found: C, 49.85; H, 6.93; N, 9.71.
3.4. Xylan derivatization
3.4.1. Hydroxypropylated xylan
Never-dried (7.5 wt % of xylan, containing 0.9% NaOH) or dried
xylans extracted, for example, from (bleached) birch pulp were
used as starting materials. The pH and/or sodium hydroxide con-
tent was adjusted to pH 11–13 and/or 0.5–2.0 M. The mixture
was first stirred for 1 h at 65 °C and then cooled down to room
temperature. Propylene oxide was added into the reaction mixture
keeping temperature at 15 °C. After the addition of propylene oxide
the reaction mixture was let to warm up slowly to room tempera-
ture and to react for 24 h. After the reaction the pH was adjusted to
6–7 with ꢁ6 M HCl. Xylan derivatives were first precipitated using
acetone (five times volume compared to the amount of water in
the reaction mixture) and purified by dialysis for two nights, first
night in running tap water and the second night in standing deion-
ized water (suitable membrane cut-off was 3500 Da), concen-
trated, and finally freeze-dried.
Figure 11. Swelling curves of HPX-BA, HPX-A, and X-BA cross-linked with (A) DAT,
(B) DAX, (C) DAA, and (D) DAG. The solid line represents cross-linking with 1 wt %
and the dashed line cross-linking with 5 wt % of DA.
The sodium hydroxide content of the solution was adjusted to
1.0 M. The mixture was first stirred for 1 h at 65 °C and then 24 h
at room temperature. The solution was heated to 45 °C under ar-
gon. Butyl glycidyl ether (64.5 mL) and allyl glycidyl ether
(28.7 mL) were mixed and added dropwise into the reaction mix-
ture and left to react for 24 h at 45 °C. The solution turned white
and turbid. After cooling the reaction mixture turned into a brown-
ish solution and the pH was adjusted to 6–7 with ꢁ6 M HCl. The
3.4.2. Derivatization of hydroxypropylated xylan with butyl and
allyl glycidyl ether
600 mL (627 g, 11.4 wt % of xylan) of hydroxypropylated xylan
solution was placed in a 1-l reactor with a mechanical stirrer.

2744
H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
product was first precipitated with 4 L of acetone, left to settle
overnight and washed with 2 ꢂ 1 L of acetone. Excess acetone
was decanted and the precipitate left to dry in air overnight. The
product was dissolved in 250 mL of water and purified by dialysis
for two nights, first night in running tap water and the second
night in standing deionized water (suitable membrane cut-off
was 3500 Da) and finally freeze-dried to obtain 26.9 g of white
crystalline solid. More product was obtained by evaporating the
acetone from the supernatant remaining from the precipitation
step. A slimy residue was obtained after pouring out the excess
water. The residue was diluted with a small amount of ethanol,
and dialyzed and freeze-dried as previously described to obtain
32.9 g of pure product. Overall yield 59.8 g (38%).
the mean degree of swelling was then calculated. The degree of
swelling was determined by the following equation:
mwet ꢀ mdry
Degree of swelling ¼
ꢃ 100%
mdry
4. Conclusions
Novel hydrogels were synthesized using bio-based starting
materials. Xylan was derivatized to different degrees of substitu-
tion of 1-allyloxy-2-hydroxy-propyl (A) groups combined with 1-
butyloxy-2-hydroxy-propyl (B) and/or hydroxypropyl (HP) groups.
The sugar diacids used as cross-linkers were derivatized chemi-
cally to N,N0-diallylaldardiamides (DA). The most important corre-
lation was observed between the cross-linking efficiency of the
allylic double bonds and the water absorbency. The water absor-
bency (up to 350%) was highest for hydrogels prepared from
HPX-A with the lowest cross-linking efficiency 18%. Contrarily,
the lowest water absorbency (about 100%) was observed for hydro-
gels prepared from X-BA with the highest cross-linking efficiency
of 50%. When the cross-linking efficiency of allylic double bonds
of xylan derivatives increases the incorporation of an additional
DA cross-linker into the hydrogel network decreases. However,
the amount of DA cross-linkers had only a small influence on the
water absorbencies. The use of DA cross-linkers in the preparation
of hydrogels had an influence on the pore structure making it more
uniform. The effect of HP and B substituents to water absorbency
was not so clear: the HP groups in the xylan backbone of HPX-A
(DSHP 0.9) and HPX-BA (DSHP 0.6) seemed to increase the water
absorbency probably due to longer side chains enabling a more
open structure. The more hydrophobic B groups in HPX-BA (DSB
0.3) and X-BA (DSB 0.8) hydrogels may also have a negative effect
on the water absorption properties.
3.4.3. Derivatization of hydroxypropylated xylan with allyl
glycidyl ether
The reaction was performed as in Section 3.4.2 but t-BuOH
(364 mL) was added as a co-solvent to the reaction mixture and
only allyl glycidyl ether (77 mL) was used. The entire product
was obtained from the precipitation step. Yield was 47.4 g (41%).
3.4.4. Derivatization of xylan with allyl and butyl glycidyl ethers
700 mL (750 g, 7.3 wt % of xylan) of a suspension containing
51.1 g of the non-dried xylan was used as a starting material. NaOH
concentration of the suspension was 0.8 M. The reaction mixture
was first heated to 65 °C for 2 h. 100 mL of butyl glycidyl ether
was slowly added (50 min) and let to react overnight. 50 mL of allyl
glycidyl ether was slowly added (50 min) and let to react for an
additional 24 h. The reaction mixture was adjusted to pH 7–8 with
37% HCl. The neutralized reaction mixture was slowly poured into
2.5 L of acetone, excess acetone was decanted, and the precipitate
left to dry in air overnight. The product was purified by dialysis.
The dialyzed xylan in water was freeze-dried yielding 76 g (40%)
of solid white powder.
Acknowledgments
3.5. Preparation of hydrogels
This work was supported by the Academy of Finland through the
Finnish Centre of Excellence in White Biotechnology-Green Chem-
istry-project and by The Finnish Funding agency for Technology and
Innovation (TEKES) through Forestcluster Ltd’s Future Biorefinery
(Fubio) Program in task hemicelluloses and hydroxy acids.
To a solution of 2.0 g of xylan derivative with or without 1 or
5 wt % of DA cross-linker in 20 mL of deionized water, 100 mg
(5 wt %) of radical initiator potassium persulfate was added. The
solution was poured onto a Petri dish, placed in an UV oven and
polymerized under UV light for 3–4 min during which the gels
turned more white and turbid. After cross-linking the gels were
washed several times with deionized water to remove any unre-
acted material and salts. The gels were dried in a vacuum oven
in 40 °C for 48 h or by freeze-drying. Before freeze-drying the
wet gels were immersed in liquid nitrogen.
References
1. Hennink, W. E.; van Nostrum, C. F. Drug Deliv. Rev. 2002, 54, 13–36.
2. Yin, Y.-L.; Prud’homme, R. K.; Stanley, F. In Harland, R. S., Prud’homme, R. K.,
Eds.; Polyelectrolyte Gels: Properties, Preparation, and Applications; ACS
Symposium Series: Washington, DC, 1992; Vol. 480, pp 91–113. Chapter 6.
3. Lee, K. Y.; Mooney, D. J. Chem. Rev. 2001, 101, 1869–1879.
4. Alvarez-Lorenzo, C.; Concheiro, A. Mini-Rev. Med. Chem. 2008, 8, 1065–1074.
5. Voepel, J.; Sjöberg, J.; Reif, M.; Albertsson, A.-C.; Hultin, U.-K.; Gasslander, U. J.
Appl. Polym. Sci. 2009, 112, 2401–2412.
6. Guilherme, M. R.; Reis, A. V.; Paulino, A. T.; Moia, T. A.; Mattoso, L. H. C.;
Tambourgi, E. B. J. Appl. Polym. Sci. 2010, 117, 3146–3154.
7. Wang, W.; Wang, A. Adv. Mater. Res. 2010, 96, 177–182.
8. Seneviratine, C. M. W. PCT Int. Appl. WO2009/036483 A1.
9. Voepel, J.; Edlund, U.; Albertsson, A.-C. J. Polym. Sci., Part A: Polym. Chem. 2009,
47, 3595–3606.
3.6. Bromination
The samples (2 g) were stirred in chloroform (68 mL) for 18 h
protected from light. All the samples were insoluble to chloroform.
A 12% solution of bromine in chloroform was freshly prepared and
10 mL was added to each reaction mixture until a slight excess of
bromine remained. The samples were stirred for 2 h and subse-
quently washed with methanol, aqueous sodium thiosulfate
(10%, w/v), water and once again with methanol. The samples were
dried in a vacuum oven in 40 °C for 24 h.
10. Tanodekaew, S.; Channasanon, S.; Uppanan, P. J. Appl. Polym. Sci. 2006, 100,
1914–1918.
11. Gabrielii, I.; Gatenholm, P. J. Appl. Polym. Sci. 1998, 69, 1661–1667.
12. Lindblad, M.; Ranucci, E.; Albertsson, A.-C. Macromol. Rapid Commun. 2001, 22,
962–967.
13. Silva, T. C. F.; Habibi, Y.; Colodette, J. L.; Lucia, L. A. Soft Matter 2011, 7, 1090–
1099.
3.7. Swelling assessments
14. Mojzita, D.; Wiebe, M.; Hilditch, S.; Boer, H.; Penttilä, M.; Richard, P. Appl.
Environ. Microbiol. 2010, 76, 169–175.
15. Van Esch, J. H.; Heeres, A. WO 02/070463 A1, 2002.
16. Anker, H. S. FEBS Lett. 1970, 7, 293.
17. Tabern, D. L. U.S. Patent 2,084,626, 1937.
18. Henkensmeier, D.; Cajus Abele, B.; Candussio, A.; Thiem, J. J. Polym. Sci., Part A:
Polym. Chem. 2005, 43, 3814.
For swelling measurements, the gel samples were dried to a
constant weight in 40 °C in a vacuum oven (mdry) and immersed
in excess of deionised water at room temperature. Excess water
was removed with a dry filter paper and the sample (mwet
)
weighed at time intervals. The gels were evaluated in triplets and

H. Pohjanlehto et al. / Carbohydrate Research 346 (2011) 2736–2745
2745
19. Huijbrechts, A. M. L.; Desse, M.; Budtova, T.; Franssen, M. C. R.; Visser, G. M.;
24. Glasser, W. G.; Jain, R. K.; Sjöstedt, M. A. U.S. Patent 5,430,142, 1995.
25. Jain, R. K.; Sjöstedt, M.; Glasser, W. G. Cellulose 2001, 7, 319–336.
26. Duanmu, J.; Gamstedt, E. K.; Rosling, A. Starch/Stärke 2007, 59, 523–532.
27. Heinze, T.; Lincke, T.; Fenn, D.; Koschella, A. Polym. Bull. 2008, 61, 1–9.
28. Lepistö, M.; Artursson, P.; Edman, P. Anal. Biochem. 1983, 133, 132–135.
29. Voepel, J.; Edlund, U.; Albertsson, A.-C. Polym. Prepr. 2010, 51, 747–748.
Boeriu, C. G.; Sudhölter, E. J. R. Carbohydr. Polym. 2008, 74, 170–184.
20. Shen, X.; Kitajyo, Y.; Duan, Q.; Narumi, A.; Kaga, H.; Kaneko, N.; Satoh, T.;
Kakuchi, T. Polym. Bull. 2006, 56, 137–143.
21. Kiely, D. E.; Carter, A.; Shrout, D. P. U.S. Patent 5,599,977, 1997.
22. Kiely, D. E.; Chen, L.; Lin, T. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 594–603.
23. Fuhrmann, A.; Krogerus, B. TAPPI Press—TAPPI Engineering, Pulping and
Environmental Conference 2009—Innovations in Energy, Fiber and
Compliance, Memphis, TN, October 2009.
30. Kabiri, K.; Omidian, H.; Hashemi, S. A.; Zohuriaan-Mehr, M. J. Eur. Polym. J.
2003, 39, 1341–1348.