1528 Biomacromolecules 2004, 5, 1528-1535

Material Properties of Plasticized Hardwood Xylans for Potential

Application as Oxygen Barrier Films
Maria Gröndahl, Lisa Eriksson, and Paul Gatenholm*
Biopolymer Technology, Department of Materials and Surface Chemistry,
Chalmers University of Technology, SE-412 96 Göteborg, Sweden
Received February 6, 2004; Revised Manuscript Received March 24, 2004

Free films based on glucuronoxylan isolated from aspen wood were prepared by casting from aqueous
solutions and drying in a controlled environment. Addition of xylitol or sorbitol facilitated film formation
and thus examination of the material properties of these films. The mechanical properties of the films were
evaluated using tensile testing and dynamic mechanical analysis in a controlled ambient relative humidity.
The strain at break increased, and the stress at break and Young’s modulus of the films decreased with
increasing amounts of xylitol and sorbitol due to plasticization. At high amount of plasticizer, it was found
that films with xylitol gave lower extensibility. Wide-angle X-ray scattering analysis showed that xylitol
crystallized in a distinct phase, which we believe contributes to the more brittle behavior of these films. The
effect of the plasticizers on the glass transition temperature was determined using dynamic mechanical
analysis and differential scanning calorimetry. An increased amount of plasticizer shifted the glass transition
to lower temperatures. The effect of moisture on the properties of plasticized films was investigated using
water vapor sorption isotherms and by humidity scans in dynamic mechanical analysis. Sorption isotherms
showed a transition from type II to type III when adding plasticizer. The films showed low oxygen
permeability and thus have a potential application in food packaging.

Introduction very good oxygen barrier properties,11-12 but a drawback of

The great majority of plastic materials in use today are
based on fossilic raw materials. More than 40% of all plastic
materials are used as packaging, and their disposal contributes
to growing landfills and enhanced greenhouse effects when
burned.1 Furthermore, the fossil resources on earth are
limited. A sustainable development in the future requires the
use of renewable materials. The use of plant biopolymers to
produce plastic materials has several advantages. In addition
to their being a renewable resource, the product may after
use be composted, recycled or incinerated with minimal
environmental impact.2
Food packaging requires materials with good mechanical
properties so that the food remains undamaged during
storage. Good barrier properties are also necessary to prevent
gases such as oxygen from degrading the product. Such
materials should furthermore be resistant to water and it is
important that they do not emit poisonous substances since
they are in contact with food.3 Starch films have been
evaluated for possible use in food packaging applications.4-7
Starch is present in large quantities in potato and maize, for
example, and has been proven to fulfill a number of
necessary criteria for food packaging applications. The
mechanical properties are excellent and the flexibility can
be further improved by the addition of plasticizers. Glycerol
is frequently added as a low molecular weight plasticizer,8-10
but xylitol and sorbitol have also been used.7 Starch exhibits
* To whom correspondence should be addressed. Fax: +46 (0)31-
7723418. E-mail: paul.gatenholm@chem.chalmers.se.
10.1021/bm049925n CCC: $27.50
Published on Web 05/11/2004
the material if used on a large scale is that it would then
compete with the food industry.
Hemicellulose may be an alternative material for such
applications. It is one of the main constituents of wood and
other plants and exists in major quantities in agricultural
waste. Approximately 60 billion tons of hemicellulose is
biosynthesized every year.13 So far, the commercial utiliza-
tion of hemicellulose in nonfood applications has been very
limited. In the pulp and paper industry, retaining the
hemicellulose in the pulp has been shown both to improve
the mechanical properties of the paper and the yield.14,15
Hemicellulose has also been used as sweetening agents,
thickeners, and emulsifiers in food.16 There has been interest
in the use of hemicellulose as a nutraceutical,17 in chiral
separations,18 and as an HIV inhibitor.19
Hemicelluloses are heteropolysaccharides whose composi-
tion varies between different plant species. In hardwoods
such as aspen, beech, and birch, however, the hemicellulose
consists chiefly of O-acetyl-(4-O-methylglucurono)xylan.20
The backbone consists of β-(1f4)-linked D-xylopyranosyl
residues substituted with one R-(1f2)-linked 4-O-methyl-
D-glucuronic acid per approximately every 10th such residue.
The xylopyranosyl residues are partially acetylated at the
C-2 and/or C-3 positions with a degree of acetylation between
0.6 and 0.7 in aspen wood.21
In many aspects, the properties of glucuronoxylan lie
between those of cellulose and starch. Glucuronoxylan is a
cell wall polymer, like cellulose, but cellulose is a linear
polymer with a rigid backbone designed to crystallize and
© 2004 American Chemical Society
Properties of Plasticized Hardwood Xylans
be loadbearing. Glucuronoxylan is, as mentioned above,
substituted with glucuronic acid and acts as a matrix material
in the cell wall. Starch, which is an energy reserve in the
plant, has an extensively branched component, amylopectin.
Cellulose and starch are semicrystalline in their native state,
whereas glucuronoxylan is amorphous in the native state but
can crystallize after isolation using alkali.22 The crystalline
structure incorporates water, as in starch.23 However, glu-
curonoxylan is not as easily soluble in water as starch,24
which could be an advantage in using it in food packaging.
The aim of this study was to evaluate glucuronoxylan
isolated from hardwood for potential application as food
packaging. This work focuses on investigating the effect of
addition of xylitol and sorbitol on film formation of glucu-
ronoxylan. The effect on the mechanical and barrier proper-
ties of the films and the interactions with water were
evaluated.
Materials and Methods
Materials. The aspen glucuronoxylan used in this study
was isolated by alkali extraction at the Department of Wood
Sciences and Forest Products at Virginia Tech, Blacksburg,
VA. The isolated material contained 83 wt. % xylose, 14
wt. % methyl glucuronic acid, 2 wt. % mannose, and less
than 1 wt. % of other sugars. The weight average molecular
weight was 15 000 g/mol, determined using size exclusion
chromatography (SEC) in DMSO. The isolation procedure
and characterization are described in detail elsewhere.22,25
Xylitol and sorbitol were purchased from Fluka (article
no X-3375) and Sigma (article no 85529), respectively.
Preparation of Films. Films were prepared by mixing
glucuronoxylan, xylitol, or sorbitol and deionized water
during magnetic stirring at 95 °C for 15 min. The total
amount of dry substance, glucuronoxylan and an additive,
in each film was kept constant at 1 g and 35 mL of water
was added. The additive content was varied between 20 and
50 wt. % of the dry weight (20, 27.5, 35, 42.5, and 50 wt.
% of xylitol and 20, 35, and 50 wt. % of sorbitol). The
solutions were poured onto polystyrene Petri dishes with a
diameter of 14 cm, and films were allowed to form upon
drying in a temperature of 23 °C and a relative humidity
(RH) of 50%. The dried films were stored in these conditions
until analysis.
Films of poly(vinyl alcohol) were prepared to compare
the oxygen permeability with that of the glucuronoxylan
films. 1 g of poly(vinyl alcohol) (Fluka 81382) was dissolved
in 35 mL of deionized water at 95 °C for 15 min during
magnetic stirring. The solutions were poured onto dishes
similar to those used for the glucuronoxylan solutions and
dried under the same conditions.
Tensile Testing. The mechanical properties of the con-
ditioned films were measured with a tensile testing machine
(Lloyd L2000R) with a load cell of 100 N capacity. The
samples were cut into dog-bone-shaped strips with a width
of 7 mm. The thickness of the samples, measured with a
micrometer, was 35 ( 5 µm. The initial distance between
the grips was 20 mm and the separation rate of the grips
was kept constant at 5 mm per minute. Fifteen replicates of
Biomacromolecules, Vol. 5, No. 4, 2004 1529
the films with 20, 35, and 50 wt. % additive and five
replicates of the films containing 27.5 and 42.5 wt. % xylitol
were tested. The stress-strain curve was recorded for each
sample, and Young’s modulus, strength at break, and strain
at break were calculated. The measurements were performed
at 50% RH and 23 °C after conditioning for at least one
week.
Wide-Angle X-ray Scattering. The crystallinity of the
materials was determined with wide-angle X-ray scattering
(WAXS). The films were milled in liquid nitrogen, and the
samples were investigated with a Siemens D5000 gonio-
metric diffractometer. Cu KR radiation with a wavelength
of 1.54 Å was used, and 2θ was varied between 5 and 30°
at a rate of 1° (2θ) per minute and a step size of 0.1° (2θ).
A relative crystallinity was calculated to compare the
crystallinity of the samples. The background was subtracted
from the diffractogram by drawing a baseline tangentially
to the curve minimum. The relative crystallinity was
determined by comparing the area of the crystalline peak at
about 18° (2θ) with the total area under the peak.26
Dynamic Mechanical Analysis. Temperature scans in
dynamic mechanical analysis (DMA) were performed in
tension at 1 Hz in a Rheometrics RSA-II (Rheometrics
Scientific, Piscataway, NJ) in order to determine the glass
transition temperature at 50% RH. 4 × 20 mm strips of the
conditioned films were covered with a hydrophobic grease
(Stabox 9415 provided by AB Axel Christiernsen in Nol,
Sweden) and heated from temperatures of -70 up to +110
°C with a heating rate of 5 °C per minute.
The mechanical behavior of the films as a function of the
surrounding relative humidity was studied with DMA
humidity scans. A Perkin-Elmer DMA7 with extension
assembly connected to a humidity controller accessory was
used. Strips of the films, 5 × 20 mm in size, were mounted
between the two holders. The samples were then loaded in
tensile mode with a frequency of 1 Hz and an amplitude of
about 4 µm. The temperature was 30 °C, and the relative
humidity in the sample chamber during testing was changed
according to a humidity program. The program was initiated
with 30 min at 1% RH, followed by a ramp of 1% RH per
minute until 90% RH was reached, and was finished with
five minutes at 1% RH. The air humidity at which the
transition took place was taken from the onset of the decrease
in the storage modulus. This is not the transition at
equilibrium water content, wg, but the relative humidity at
which softening takes place under the described conditions,
providing a comparison between the films.
Differential Scanning Calorimetry. The thermal transi-
tions of the conditioned films were studied using differential
scanning calorimetry (DSC). The instrument used was a
Perkin-Elmer Pyris. Pieces of the RH conditioned films were
heated from -110 to +110 °C in sealed steel cups with
heating and cooling rates of 10 °C per minute. All measure-
ments were made in at least triplicate. The inflection point
of the heat flow step change, close to the half Cp change of
the second heating scan was taken as the operative definition
of the glass transition temperature Tg. Before and after the
measurements, the cups were weighed in order to ensure that
no evaporation had taken place.
1530 Biomacromolecules, Vol. 5, No. 4, 2004 Gröndahl et al.

Water Vapor Sorption Isotherms. Pieces of the films

were conditioned in closed vessels containing saturated salt
solutions. The salts used were lithium chloride (Sigma-
Aldrich R13013), magnesium chloride hexahydrate (Sigma-
Aldrich R31413), magnesium nitrate hexahydrate (Sigma-
Aldrich F63079), sodium chloride (Sigma-Aldrich F71381),
and potassium sulfate (Sigma-Aldrich R31270), resulting in
relative humidities of 11, 33, 54, 75, and 97%, respectively,
at ambient temperature according to ASTM E104-85. The
films were first conditioned at the lowest relative humidity
and then moved successively to the highest relative humidity.
Finally, the samples were dried at 130 °C overnight. The
equilibrium water content was measured gravimetrically on
a Mettler AE260-DR balance and calculated as the weight
of water in the sample at equilibrium compared to the total
weight. The points displayed are average values calculated
from triplicates.
Oxygen Permeability. The oxygen transmission of the
films was measured in accordance with ASTM D3985-81
using Mocon Oxtran 2/20 equipment (Modern Controls Inc.,
Minneapolis, MN) with a coulometric oxygen sensor. The
area of the samples was 5 cm2, and the analysis was made
in 50% RH after conditioning of the samples. Data were
collected for 24 h. The permeability was calculated from
the transmission and the measured thickness of the films and
is presented in units of (cm3 µm)/(m2 d kPa), where 1 d )
24 h.

Results and Discussion

Film Formation. Films prepared from glucuronoxylan

alone became very brittle and fragmented upon drying. Film
formation conditions, such as temperature and surrounding
humidity, were kept constant and varying the drying time
did not result in any change of the film forming ability.
Possible explanations of the poor film forming ability would
be either insufficient chain length of the polymer, high glass
transition temperature, or poor solubility. We have seen that
slight carboxymethylation of this very aspen glucuronoxylan
improves film formation.22 Thus, the molecular weight
requirements should be fulfilled. The glass transition tem-
perature of hardwood xylan is difficult to determine due to
thermal degradation at elevated temperatures. The decom-
position temperature of hemicellulose is lower than that of
cellulose and lignin27-28 but can be increased by esterifi-
cation.29-30 The glass transition temperature has been esti-
mated to be around 180 °C,31-32 which might explain the
poor film formation at room temperature. To suppress the
glass transition temperature, we have added xylitol and
sorbitol as potential plasticizers. After the addition of xylitol
or sorbitol in the amounts described in the experimental part,
transparent films with good mechanical cohesion were
formed, which was a prerequisite for the evaluation of
material properties.
Mechanical Properties. The evaluation of material prop-
erties started off with the determination of tensile properties
at room temperature and 50% RH. Parts a and b of Figure
1 show typical stress-strain curves for the glucuronoxylan
films with various concentrations of xylitol and sorbitol,
Figure 1. a. Stress-strain behavior of xylitol-plasticized films at 23
°C and 50% RH, from left to right: 20, 35, and 50 wt. % xylitol. b.
Stress-strain behavior of sorbitol-plasticized films at 23 °C and 50%
RH, from left to right: 20, 35, and 50 wt. % sorbitol.
respectively. Films with 20 wt. % xylitol or sorbitol are
strong, with a stress at break above 40 MPa, but are brittle,
with an elongation at break of around 2%. The strength and
brittleness of these films are in agreement with tensile testing
results of the hydroxypropyl derivative of xylan.33 The
addition of more xylitol or sorbitol results in reduced strength
but an increase in elongation at break. Pay attention to the
difference in elongation at break with increasing amount of
sorbitol compared to xylitol, which contributes to a larger
work of fracture of the sorbitol-plasticized films.
Parts a-c in Figure 2 illustrate how the amount of
plasticizer affects the stress at break, Young’s modulus, and
strain at break of the films. The effect on strength at break
and Young’s modulus with increasing amount of plasticizer
is similar for xylitol- and sorbitol-containing films. However,
as seen in Figure 2c, the xylitol-plasticized films are less
extensible than the films plasticized with sorbitol when 50
wt. % of plasticizer is added.
Wide-Angle X-ray Scattering. The morphology of the
films was investigated using WAXS. The diffractograms of
the xylitol-containing films and xylitol itself are shown in
Properties of Plasticized Hardwood Xylans
Figure 2. a. Effect of the amount of plasticizer on the stress at break
at 23 °C and 50% RH, b, xylitol; O, sorbitol. The error bars represent
the standard deviation. b. Effect of the amount of plasticizer on
Young’s modulus at 23 °C and 50% RH, b, xylitol; O, sorbitol. The
error bars represent the standard deviation. c. Effect of the amount
of plasticizer on the strain at break at 23 °C and 50% RH, b, xylitol;
O, sorbitol. The error bars represent the standard deviation.
Figure 3a and the corresponding data for the sorbitol-
containing films is shown in Figure 3b. The diffractograms
Biomacromolecules, Vol. 5, No. 4, 2004 1531
Figure 3. a. X-ray diffractograms of samples containing xylitol, from
bottom to top: 0 (pure glucuronoxylan), 20, 27.5, 35, 42.5, 50, and
100 (pure xylitol, rescaled with the factor 1/3) wt. % xylitol. The
diffractograms are displaced in the y direction. b. X-ray diffractograms
of samples containing sorbitol, from bottom to top: 0, 20, 35, 50,
and 100 (rescaled with the factor 1/2) wt. % sorbitol. The diffracto-
grams are displaced in the y direction.
of the pure low molecular weight plasticizers give rise to
sharp distinct peaks, whereas the polymer has broader peaks
and a halo, resulting from scattering from the amorphous
parts.
Xylan crystallizes in a helical structure with a 3-fold screw
axis along the chain direction. The crystallinity and the
interplanar distances increase with an increasing amount of
water. The hydrated form has a layer spacing of about 14.84
Å.34-35 In the left-handed structure, an intramolecular
hydrogen bond O5‚‚‚O3′ is possible, whereas there is a steric
conflict for a right-handed helix between O2 and H4′.36 The
O5‚‚‚O3′ intramolecular hydrogen bond is weaker in xylan
than in cellulose, and the xylan helix is stabilized mainly by
van der Waal forces. The absence of strong intramolecular
hydrogen bonds makes the xylan less rigid than cellulose.37
There are six monomer units in two antiparallel chains and
six water molecules in the unit cell of the hydrated form.
The water molecules probably form a column. The empty
lattice site is able to accommodate water as well as side
1532 Biomacromolecules, Vol. 5, No. 4, 2004 Gröndahl et al.

groups such as glucuronic acid. Xylan seems to depend on

the presence of water to stabilize the structure.38 The O2-H
and O3 atoms of each xylose residue point toward one of
the unoccupied lattice sites in the structure, the expected
locations of the water molecules. In the proposed water chain,
each water molecule accepts from the xylose O2-H and
donates to an O3. The same water molecule also accepts and
donates to its neighboring water molecules, giving two strong
hydrogen bonds in addition to the weak intramolecular
hydrogen bond O3′‚‚‚O5.23
The crystalline peak at about 2θ ) 18° can be used to
compare the relative crystallinity of the samples. The addition
of plasticizer did not affect the degree of crystallinity of the
films, which was calculated as the area under the crystalline
peak at about 2θ ) 18° divided by the total area at those
angles (amorphous and crystalline scattering). The effect of
the plasticizer on the crystallinity is complex and involves
two opposite factors. On one hand, the plasticizer increases
the mobility of the polymer chains and therefore the ability
to crystallize. On the other hand, an increase in the amount
of plasticizer leads to more interactions between the plasti-
cizer and the glucuronoxylan and therefore to a decrease in
the possible interactions between the glucuronoxylan chains.
The relative crystallinity was at the same level for all films,
between 44 and 47%.
The crystalline peak at 2θ ) 18° is displaced in the
diffractograms in the plasticized samples as compared to the
pure glucuronoxylan. The distance between the lattice planes,
d, can be related to the Bragg angle, θ, using Bragg’s law:
2 d sin θ ) nλ, where n is an integer and λ is the wavelength
of the X-ray radiation. A lower angle correlates to a larger
distance between the lattice planes. The spacing between the
crystal lattice planes of the glucuronoxylan increases when
plasticizer is added, and large amounts of plasticizer give
rise to greater separation. It seems as though the low
molecular weight plasticizer molecules can penetrate the
crystalline parts of the glucuronoxylan.
Small peaks from crystalline xylitol can be observed in
the films containing 42.5 and 50 wt. % xylitol. This indicates
crystallization of xylitol in a distinct phase. The crystallized
xylitol does not plasticize the glucuronoxylan, which may
contribute to the lower elongation at break of the films with
50 wt. % xylitol compared to films with a corresponding
amount of sorbitol, observed by tensile testing. Furthermore,
the crystallized xylitol can induce fracture of the films.
Dynamic Mechanical Analysis. Temperature scans in
DMA were carried out at 50% RH. During heating, the films
lose water and become brittle. Shrinking of the films was
also observed upon evaporation of water. To prevent this,
the films were sealed with a hydrophobic grease that has
been carefully studied and shown to contribute to the
modulus only to a small extent and to give a tight encapsula-
Figure 4. a. DMA heating scan of film with 20 wt. % xylitol at 50%
tion.39 The films with 20 wt. % plasticizer have a transition RH. b Storage modulus (E′); O, loss modulus (E′′); 1, tan δ. b. DMA
above room temperature at about 40-50 °C, see Figure 4a, heating scan of film with 35 wt. % xylitol at 50% RH. b, Storage
which correlates well with the high Young’s modulus and modulus (E′); O, loss modulus (E′′); 1, tan δ. c. DMA heating scan of
low elongation at break observed during tensile testing. film with 50 wt. % xylitol at 50% RH. b, Storage modulus (E′); O,
loss modulus (E′′); 1, tan δ.
Transitions are observed as a rapid decrease in storage
modulus and a peak in loss modulus and phase angle (tan ably lower transition temperatures at 50% RH, see Figure
δ). The films with higher plasticizer content have consider- 4, parts b and c. Films with 35 wt. % xylitol have a transition
Properties of Plasticized Hardwood Xylans
Figure 5. DSC second heating scans of the plasticized films. From
bottom to top: film with 20 wt. % sorbitol, 35 wt. % sorbitol, 50 wt. %
sorbitol, 20 wt. % xylitol, 35 wt. % xylitol, and 50 wt. % xylitol. The
thermograms are displaced in the y direction.
at about -30 °C and films with 50 wt. % xylitol at -45 °C.
The storage modulus at room temperature corresponds well
with data obtained during tensile testing for all samples. The
reason why the modulus in the rubbery plateau is still in the
range of 1 Gpa is that it is a semicrystalline material. The
high values of the modulus at -50 to -60 °C is due to that
the material is frozen.
Differential Scanning Calorimetry. DSC was used for
further study of the thermal transitions of the glucuronoxy-
lans. The measurements were performed at 50% RH using
sealed cups. Figure 5 shows that when the plasticized
glucuronoxylan films are heated a glass transition is detected
as an increase of the heat flow to a higher level. The glass
transition detected in the second heating scan was at the same
level as in the first heating scan for all samples. The glass
transition occurs below room temperature for all plasticized
films and there is a tendency toward a decrease of the
calorimetric glass transition temperature with increasing
amount of plasticizer. The film with 35 wt. % xylitol has a
transition at -35 °C, the film with 50 wt. % xylitol at -43
°C, the film with 35 wt. % sorbitol at -26 °C, and the film
with 50 wt. % sorbitol at -31 °C. This is in good agreement
with the transitions observed during DMA heating scans.
Furthermore, a melting endotherm was observed in the
thermogram of the film with 50 wt. % xylitol, supporting
the presence of crystallized xylitol. However, we cannot
explain the low glass transitions detected for the films with
20 wt. % of plasticizer.
Effect of Water. It is well-known that the properties of
polysaccharides are affected by water. We have therefore
performed our material investigations in a controlled envi-
ronment at 50% RH. The ternary system polymer-plasticizer-
water is very complex and thus deserves to be a subject for
extensive study. A first step is to examine the effect of the
surrounding humidity on the water content. We have
measured the water vapor sorption isotherms, and they are
shown in Figure 6. The standard deviations were small.
Water vapor sorption isotherms of pure hemicellulose have
Biomacromolecules, Vol. 5, No. 4, 2004 1533
Figure 6. Water sorption isotherms at 20 °C for glucuronoxylan films
with different amounts of xylitol. b, Pure glucuronoxylan; O, 20 wt.
%; 1, 27.5 wt. %; 3, 42.5 wt. %.
been reported previously.29,40 The shape of the curve for pure
xylan is characteristic for systems with strong polymer-
polymer and polymer-solvent interactions (type II iso-
therm).41 The shape arises from partial replacement of
polymer-polymer hydrogen bonds with polymer-water
hydrogen bonds. The equilibria involve bound and unbound
water and free and associated polymer. The addition of
plasticizer results in a transition from a type II to a type III
isotherm, which is typical for plasticized systems. This
difference in shape has also been observed for amylose and
amylopectin films with or without glycerol.12 It is interesting
to note that all plasticized films had a similar water content
at 50% RH at which the mechanical tests were carried out.
The plasticized films have a higher water content than the
unplasticized films at ambient RH above 70%.
Humidity scans in DMA show how the modulus of the
materials depends on the air humidity. A softening of the
material occurs at a certain RH, which leads to a rapid
decrease in the storage modulus, see Figure 7. The films
with higher amounts of plasticizer have a lower onset RH
for the decrease in modulus. The same trend was observed
in the sorbitol-plasticized films. When the relative humidity
rises, the materials take up water. Water incorporated in the
crystalline regions does not affect the mechanical properties,
whereas water present in the amorphous segments acts as a
plasticizer and reduces the stiffness of the material. A greater
amount of plasticizer reduces the intermolecular interactions
between the glucuronoxylan chains, thus facilitating interac-
tions between water and the hydroxyl groups of the glucu-
ronoxylan.
Oxygen Permeability. Polysaccharides are generally good
oxygen barriers because hydrogen bonds contribute to good
packing of the material and thus a low permeability. The
oxygen transmission of the glucuronoxylan film with 35 wt.
% sorbitol was measured at 50% RH. The average oxygen
permeability was 0.21 (cm3 µm)/(m2 d kPa). The measured
oxygen permeability is lower than literature values for
glycerol-plasticized starch and comparable with the often
1534 Biomacromolecules, Vol. 5, No. 4, 2004
Table 1. Oxygen Permeability Data on Plasticized Glucuronoxylan Film and Comparable Values for Plasticized Starch Polymers, Ethylene
Vinyl Alcohol, and Low Density Polyethylene
oxygen permeability
material [(cm3 µm)/(m2 d kPa)]
glucuronoxylan with 35 wt. % sorbitol
amylose with 40 wt. % glycerol
amylopectin with 40 wt. % glycerol
poly(vinyl alcohol) (PVA)
ethylene vinyl alcohol (EVOH)
low-density polyethylene (LDPE)
Figure 7. DMA of xylitol-plasticized films, humidity scans between
0 and 90% RH. From left to right: 50, 42.5, 35, 27.5, and 20 wt. %
xylitol.
used barrier plastic ethylene vinyl alcohol (EVOH), see Table
1.
Conclusions
Plasticization of glucuronoxylan with xylitol or sorbitol
enables the formation of transparent, continuous, and flexible
films by lowering the glass transition temperature, determined
by using DMA and DSC.
The stress at break and Young’s modulus of the films
decreased, whereas the strain at break increased with
increasing plasticizer content. Sorbitol and xylitol gave rise
to films exhibiting similar mechanical properties, but the
xylitol plasticized films were slightly less extensible at high
plasticizer levels.
Analysis with WAXS showed that the glucuronoxylan in
all samples was semicrystalline. The presence of peaks from
crystalline xylitol in the samples with large amounts of xylitol
as well as the melting endotherm in the DSC heating scan
revealed that xylitol crystallized in a distinct phase. The
crystallized xylitol does not plasticize the glucuronoxylan
but can induce fracture of the films. This might explain the
difference in strain at break between the films with 50 wt.
% xylitol and 50 wt. % sorbitol. The xylitol and sorbitol
molecules are also possibly intercallants in the crystalline
lattice.
The amount of plasticizer in the films affected the water
content of the materials. The addition of plasticizer resulted
in a transition from type II to type III water vapor sorption
Gröndahl et al.
source and conditions
0.21 present study, 50% RH
7 42 50% RH
14 42 50% RH
0.21 present study, 50% RH,
0.1-12 43 70% VOH, 0-95% RH
1870 43 50% RH
isotherms. The storage modulus of the plasticized films
decreased when the surrounding RH was increased in a
humidity scan in DMA. The higher the amount of plasticizer,
the lower the onset RH for the decrease in modulus.
The films with 35 wt. % sorbitol exhibited excellent
oxygen barrier properties at 50% RH. The oxygen perme-
ability was lower than that of plasticized starch and in the
same range as that of the commercially used synthetic
polymeric material EVOH. Thus, glucuronoxylan has a
potential in film applications, such as food packaging.
Acknowledgment. This work was funded by Vinnova,
the Swedish agency for innovation systems, through the
Prohem program. Ingela Gangby at SIK, the Swedish institute
for food and biotechnology, is acknowledged for her as-
sistance in the oxygen transmission measurements and Anne-
Mari Olsson and Lennart Salmén at STFI, the Swedish pulp
and paper research institute, as well as Mats Stading at SIK
for letting us use their DMA equipment.
References and Notes
(1) www.apme.org (the Association of Plastics Manufacturers in Europe).
(2) Gross, R. A.; Kalra, B. Science 2002, 297, 803-807.
(3) Kaplan, D. L.; Mayer, J. M.; Ball, D.; McCassie, J.; Allen, A. L.;
Stenhouse, P. In Biodegradable polymers and packaging; Thomas,
E., Ed.; Technomic Publishing Company, Inc.: Lancaster, PA, 1993;
pp 1-42.
(4) Lourdin, D.; Coignard, L.; Bizot, H.; Colonna, P. Polymer 1997, 38,
5401-5406.
(5) Rindlav, Å.; Hulleman, S. H. D.; Gatenholm, P. Carbohydr. Polym.
1997, 34, 25-30.
(6) Forssell, P.; Lahtinen, R.; Lahelin, M.; Myllärinen, P. Carbohydr.
Polym. 2002, 47, 125-129.
(7) Mathew, A. P.; Dufresne, A. Biomacromolecules 2002, 3, 1101-
1108.
(8) Beutel, H. Farbe Lack 1936, 65.
(9) Tomka, I. PCT Pat. Appl. WO 90/05161, 1990.
(10) Shogren, R. L.; Swanson, C. L.; Thompson, A. R. Starch/Stärke 1992,
44, 335-338.
(11) Mark, A. M.; Roth, W. B.; Mehltretter, C. L.; Rist, C. E. Food
Technol. 1966, 20, 75-77.
(12) Stading, M.; Rindlav-Westling, Å.; Gatenholm, P. Carbohydr. Polym.
2001, 45, 209-217.
(13) Hemicelluloses: Science and Technology; Gatenholm, P., Tenkanen,
M., Eds.; Oxford University Press: New York, 2004.
(14) Meller, A. Holzforschung 1965, 19, 118-124.
(15) Molin, U.; Teder, A. Nord. Pulp Pap. Res. J. 2002, 17, 14-19.
(16) Popa, V. I. In Polysaccharides in medicinal applications; Dumitriu,
S., Ed.; Dekker: New York, 1996; pp 107-124.
(17) Sugawara, M.; Suzuki, T.; Totsuka, A.; Takeuchi, M.; Ueki, K.
Starch/Stärke 1994, 46, 335-337.
(18) Okamoto, Y.; Kawashima, M.; Hatada, K. J. Am. Chem. Soc. 1984,
106, 5357-5359.
(19) Magerstaedt, M.; Meichsner, C.; Schlingmann, M.; Schrinner, E.;
Walch, A.; Wiesner, M.; Winkler, I.; Bader, H.; Paessens, A. DE
3,921,761, 1991.
(20) Timell, T. E. Wood Sci. Technol. 1967, 1, 45-70.
(21) Teleman, A.; Lundqvist, J.; Tjerneld, F.; Stålbrand, H.; Dahlman,
O. Carbohydr. Res. 2000, 329, 807-815.
Properties of Plasticized Hardwood Xylans
(22) Gustavsson, M.; Bengtsson, M.; Gatenholm, P.; Glasser, W. G.;
Teleman, A.; Dahlman, O. In Biorelated PolymerssSustainable
Polymer Science and Technology; Chiellini, H. G. E., Braunegg, G.,
Buchert, J., Gatenholm, P., van der Zee, M., Eds.; Kluwer Academic/
Plenum Publishers: New York, 2001; pp 41-52.
(23) Nieduszynski, I. A.; Marchessault, R. H. Biopolymers 1972, 11,
1335-1344.
(24) Ebringerová, A.; Hromádková, Z. Biotechnol. Gen. Eng. ReV. 1999,
16, 325-346.
(25) Gabrielii, I.; Gatenholm, P.; Glasser, W. G.; Jain, R. K.; Kenne, L.
Carbohydr. Polym. 2000, 43, 367-374.
(26) Hermans, P. H.; Weidinger, A. J. Appl. Phys. 1948, 19, 491-
506.
(27) Ramiah, M. V. J. Appl. Polym. Sci. 1970, 14, 1323-1337.
(28) Shafizadeh, F.; Chin, P. P. S. ACS Symp. Ser. 1977, 43, 57-81.
(29) Gröndahl, M.; Teleman, A.; Gatenholm, P. Carbohydr. Polym. 2003,
52, 359-366.
(30) Buchanan, C. M.; Buchanan, N. L.; Debenham, J. S.; Gatenholm,
P.; Jacobsson, M.; Shelton, M. C.; Watterson, T. L.; Wood, M. D.
Carbohydr. Polym. 2003, 52, 345-357.
(31) Back, E. L.; Salmén, N. L. Tappi 1982, 65, 107-110.
(32) Irvine, G. M. Tappi 1984, 67, 118-121.
Biomacromolecules, Vol. 5, No. 4, 2004 1535
(33) Jain, R. K.; Sjöstedt, M.; Glasser, W. G. Cellulose 2000, 7, 319-
336.
(34) Marchessault, R. H.; Liang, C. Y. J. Polym. Sci. 1962, 59, 357-
378.
(35) Marchessault, R. H.; Settineri, W. J. Polym. Lett. 1964, 2, 1047-
1051.
(36) Settineri, W. J.; Marchessault, R. H. J. Polym. Sci., Part C 1965,
11, 253-264.
(37) Sundararajan, P. R.; Rao, V. S. R. Biopolymers 1969, 8, 305-312.
(38) Nieduszynski, I.; Marchessault, R. H. Nature (London) 1971, 232,
46-47.
(39) Stading, M. Annu. Trans. Nord. Rheol. Soc. 1998, 6, 147-150.
(40) Christensen, G. N.; Kelsey, K. E. Aust. J. Appl. Sci. 1958, 9, 265-
282.
(41) Sala, R. M.; Tomka, I. A. Angew. Makromol. Chem. 1992, 199, 45-
63.
(42) Rindlav-Westling, Å.; Stading, M.; Hermansson, A.-M.; Gatenholm,
P. Carbohydr. Polym. 1998, 36, 217-224.
(43) McHugh, T. H.; Krochta, J. M. In Edible coatings and films to
improVe food quality; Nisperos-Carriedo, M. O., Ed.; Technomic
Publishing Company: Lancaster, PA, 1994; pp 139-187.
BM049925N