PROPERTIES OF FLEXIBLE FILMS MADE OF HEMICELLULOSE
By
Lei Wang

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Packaging—Doctor of Philosophy
2015

ABSTRACT
PROPERTIES OF FLEXIBLE FILMS MADE OF HEMICELLULOSE
By
Lei Wang
Hemicellulose is the second most abundant component of plant cell walls after cellulose.
It has received increasing interest as an alternative to petroleum based polymers for packaging
applications because of its abundance, renewability and biodegradability. However, there are
some challenges in the utilization of hemicellulose as a packaging material. Hemicelluloses are
complex heteropolysaccharides with varied compositions and structures depending on the
hemicellulose source (e.g., plant species, location and growth stage); they can also be influenced
by the chemical, physical and biological isolation processes used to isolate them, and by any
pretreatment used before isolation. It is critical to understand the influence of these factors
(hemicellulose source, isolation approach, and pretreatment) on the final hemicellulose
properties, such as mechanical and barrier properties, which are important for packaging
applications. In addition, pure hemicellulose obtained from plants is water sensitive, brittle and
weak, making it unsuitable for most packaging applications even though it is a good barrier for
oxygen. Modification of hemicellulose is needed to meet the requirements of packaging
materials such as protection against chemical, mechanical, physical and biological damage, and
good barrier against water, oxygen, odors, etc.
This research focuses on the fabrication and characterization of hemicellulose based films
with adequate properties to enable their potential application as packaging materials. Three types
of raw materials, aspen ChemiThermoMechanical Pulp (CTMP), spruce CTMP and hybrid
poplar powder, with different pretreatments (bleaching and no bleaching) were employed to

obtain hemicelluloses by an alkaline extraction method. The extracted hemicellulose was further
formed into film by solvent casting in water with the addition of sorbitol and glutaraldehyde.
These two chemicals were used as a plasticizer and a cross-linking agent, respectively, to
improve the properties of hemicellulose films.
Mechanical and barrier properties of the hemicellulose based films were determined.
Hemicellulose sources and pretreatments played an important role in the properties of
hemicellulose based films. Hemicellulose based films from less treatment (powder samples, no
bleaching) displayed better properties than the films from more treatment (pulp samples,
bleaching) in general. Hemicellulose based films prepared from hybrid poplar powder without
bleaching exhibited the highest tensile strength (76.5 ± 7.0 MPa) and elongation at break (5.4 ±
0.9%), which were higher than the values obtained from other studies of similar hemicellulose
based films. Glutaraldehyde was suggested to be responsible for this improvement. The modulus
of elasticity (MOE) (3.5 ± 0.4 GPa) was comparable with those obtained for similar
hemicellulose based films in other studies. The differences in molecular weight, lignin content
and crystallinity for the hemicellulose film obtained from different approaches were believed to
result in the differences in tensile properties. No significant improvement in the barrier properties
(water vapor and oxygen barrier properties) of the hemicellulose based films was observed in
this study, which is similar to the results obtained in other studies. Hemicellulose based films
prepared from both unbleached spruce CTMP and hybrid poplar powder had better barrier
properties than the other films as indicated by the lower water vapor transmission rate (WVTR)
values obtained at 23 °C and 100% RH, but not as good as most current commercial packaging
films whose WVTR values were several orders of magnitude lower.

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to all of the people who have ever given
me your help and encouragement on my way of pursuing the Ph.D. degree during the past five
years. Without your support, this research could not be completed in time.
First, I would like to give my sincere gratitude to my major professor, Dr. D. Pascal
Kamdem. Without his strong support, guidance, and especially patience, I couldn’t finish this
program. Not only the way of his thinking to develop a project that I tried to learn, but also some
―small things‖, like dealing with the individual datum, ―In research, no bad data‖. I greatly
appreciate the invaluable suggestion, support and help from my dear guidance committee
members, Dr. Susan E. M. Selke, Dr. Bruce Harte, and Dr. David Hodge. Your kindness and
warm encouragement are engraved on my heart.
Additionally, I would like to thank Dr. Richard J. Staples for the help on the X-ray
diffractometer, Dr. Carol S. Flegler for the help on the Scanning electron microscope, Dr. Dhruv
Sharma and Abhishek Kaul for the help on the statistical analysis and also thank Dr. Ryan
Stoklosa, Muyang Li, Daniel Williams and Jacob Crowe for all the help on the high performance
liquid chromatography, size exclusion chromatography and enzyme hydrolysis. I would like to
thank all the faculty and staff, my colleagues at school of packaging and friends in and outside of
school, who have ever given me the warm encouragement, great help and moral support.
Financial support from Chinese Scholarship Council and school of packaging were also
gratefully acknowledged.
At last but not least, my deepest gratitude goes to my dearest family members for your
endless love and unselfish dedication.

iv

TABLE OF CONTENTS

LIST OF TABLES ..................................................................................................................... viii
LIST OF FIGURES ...................................................................................................................... x
1 BACKGROUND AND MOTIVATION .................................................................................. 1
1.1 Introduction and research motivation ................................................................................... 1
1.2 Objectives ............................................................................................................................. 3
2 LITERATURE REVIEW ......................................................................................................... 4
2.1 Introduction ........................................................................................................................... 4
2.2 Biodegradable polymers from renewable resources currently used in packaging................ 5
2.2.1 Cellulose ........................................................................................................................ 5
2.2.2 Starch ............................................................................................................................. 7
2.2.3 Protein .......................................................................................................................... 10
2.2.4 Other biopolymers ....................................................................................................... 10
2.3 Hemicellulose ..................................................................................................................... 11
2.3.1 Hemicellulose sources ................................................................................................. 12
2.3.1.1 Wood sources ........................................................................................................ 12
2.3.1.1.1 The state of wood resources ........................................................................... 12
2.3.1.1.2 Internal environment of hemicellulose in wood ............................................ 14
2.3.1.1.2.1 Cellulose ................................................................................................. 14
2.3.1.1.2.2 Lignin ...................................................................................................... 15
2.3.1.1.2.3 Hemicellulose ......................................................................................... 19
2.3.1.2 Pulp source ............................................................................................................ 21
2.3.2 Hemicellulose definition and utilization ...................................................................... 22
2.4 Problems for exploitation of hemicellulose in flexible packaging films ............................ 23
2.5 Review of studies on hemicellulose for packaging applications ........................................ 25
2.5.1 Different sources for hemicellulose and their compositions and structures ................ 25
2.5.2 Isolation methods ......................................................................................................... 25
2.5.2.1 Steam treatment .................................................................................................... 26
2.5.2.2 Microwave treatment ............................................................................................ 26
2.5.2.3 Explosion treatment .............................................................................................. 26
2.5.2.4 Hemicelluloses from process water ...................................................................... 27
2.5.2.5 Alkaline extraction ................................................................................................ 28
2.5.3 Influence of composition and structure on the final product properties ...................... 28
2.5.4 Modification methods .................................................................................................. 29
2.5.4.1 Addition of plasticizer and cross-linking agents ................................................... 29
2.5.4.2 Esterification and etherification ............................................................................ 30
2.5.4.3 Surface grafting ..................................................................................................... 31
2.5.4.4 Lamination ............................................................................................................ 31
2.5.4.5 Surface coating...................................................................................................... 32
2.5.4.6 Composites............................................................................................................ 32
2.5.4.7 Enzymatic modification ........................................................................................ 33

v

2.6 Economic evaluation for hemicellulose application ........................................................... 33
3 MATERIALS AND METHODS ............................................................................................ 34
3.1 Introduction ......................................................................................................................... 34
3. 2 Materials and chemicals..................................................................................................... 37
3.3 Fabrication of xylan based films by solvent casting ........................................................... 37
3.3.1 Function of additives ................................................................................................... 37
3.3.2 Proper amount of additives .......................................................................................... 38
3.4 Hemicellulose extraction and film fabrication .................................................................... 38
3.4.1 Extractives removal ..................................................................................................... 38
3.4.2 Metal ion chelation ...................................................................................................... 39
3.4.3 Bleaching ..................................................................................................................... 40
3.4.4 Alkaline extraction ....................................................................................................... 40
3.4.5 Hemicellulose recovery ............................................................................................... 41
3.4.6 Fabrication of extracted hemicellulose films by solvent casting ................................. 41
3.5 Characterization and property analysis ............................................................................... 41
3.5.1 HPLC analysis ............................................................................................................. 41
3.5.2 SEC measurement ........................................................................................................ 42
3.5.3 DP measurement by reducing-end method .................................................................. 44
3.5.4 Enzyme hydrolysis of hemicellulose ........................................................................... 45
3.5.5 FTIR analysis ............................................................................................................... 46
3.5.6 UV-Vis analysis ........................................................................................................... 47
3.5.7 XRD analysis ............................................................................................................... 47
3.5.8 DSC analysis ................................................................................................................ 48
3.5.9 Tensile tests.................................................................................................................. 48
3.5.10 SEM analysis ............................................................................................................. 51
3.5.11 Water vapor barrier property test ............................................................................... 52
3.5.12 Oxygen barrier property test ...................................................................................... 53
3.6 Statistical analysis ............................................................................................................... 53
4 PROPERTIES OF XYLAN BASED FILMS AND EXTRACTED HEMICELLULOSE
BASED FILMS............................................................................................................................ 55
4.1 Xylan based films ............................................................................................................... 55
4.1.1 Function of additives ................................................................................................... 55
4.1.2 Proper amount of additives .......................................................................................... 55
4.2 Extracted hemicellulose based films................................................................................... 56
4.2.1 UV-Vis spectra of hemicelluloses based films ............................................................ 56
4.2.2 Thermal properties of hemicellulose based films ........................................................ 59
4.2.3 Mechanical properties of hemicellulose based films ................................................... 62
4.2.4 Morphology of hemicellulose based films surface ...................................................... 72
4.2.5 Water vapor barrier properties of hemicellulose based films ...................................... 72
4.2.6 Oxygen barrier properties of hemicellulose based films ............................................. 76
5 FACTORS THAT INFLUENCE THE PROPERTIES OF EXTRACTED
HEMICELLULOSE BASED FILMS ....................................................................................... 79
5.1 Composition of extracted hemicellulose samples ........................................................... 79

vi

5.2 Molecular weight information analysis of hemicellulose samples ................................. 81
5.3 Molecular weight information of purchased xylans & HPPB hemicellulose samples and
the tensile properties of their films ....................................................................................... 89
5.4 Lignin identification and semi-quantification ................................................................. 92
5.5 Crystallinity of hemicellulose based films ...................................................................... 96
5.6 Influence of additives on the tensile properties of hemicellulose films........................ 100
5.7 Discussion ..................................................................................................................... 102
6 CONCLUSION ...................................................................................................................... 105
7 FUTURE WORK ................................................................................................................... 107
APPENDIX ................................................................................................................................ 109
REFERENCES .......................................................................................................................... 136

vii

LIST OF TABLES

Table 1 Examples of commercially utilized biopolymers. Adapted from Martín-Closas and
Pelacho (2011). ....................................................................................................................... 2
Table 2 Mechanical properties of hemicellulose films from different treatments ........................ 65
Table 3 Tensile properties of several commercially used plastics in packaging .......................... 66
Table 4 Comparison of mechanical properties of hemicellulose films with different thickness .. 71
Table 5 Water vapor permeability of hemicellulose based films from six different approaches . 75
Table 6 Water vapor transmission rates of common used plastic films ....................................... 76
Table 7 Oxygen permeability of hemicellulose based films from different approaches .............. 78
Table 8 Composition of hemicellulose samples ........................................................................... 80
Table 9 Molecular weight averages of extracted hemicelluloses specimens and their distributions
............................................................................................................................................... 88
Table 10 Comparison of purchased xylans and hemicellulose sample from HPPB and their films
............................................................................................................................................... 91
Table 11 Mechanical properties of HPP bleached hemicellulose films ..................................... 101
Table 12 Mechanical properties of HPP unbleached hemicellulose films ................................. 101
Table 13 Correlations of tensile strength, elongation and MOE ................................................ 110
Table 14 Multivariate Tests for assessing the impact of treatment on tensile properties ........... 110
Table 15 Univariate tests for assessing the impact of treatment on tensile properties ............... 111
Table 16 Multiple Comparisons among different treatment levels by Tukey test ...................... 112
Table 17 Multivariate Tests ........................................................................................................ 115
Table 18 Tests of Between-Subjects Effects .............................................................................. 116
Table 19 Multivariate Tests ........................................................................................................ 117
Table 20 Tests of Between-Subjects Effects .............................................................................. 118

viii

Table 21 MOE calculation by three different approaches .......................................................... 135

ix

LIST OF FIGURES

Figure 1 Schematic representation of cellulose molecule. Figure was adapted from Azizi Samir et
al. (2005) (Azizi Samir et al., 2005). ...................................................................................... 6
Figure 2 Amylose (a) and amylopectin (b) structures. Adapted from Carvalho (2013). ................ 8
Figure 3 Cellulose structures in trees from logs to molecules. Adapted from Dodson (2012)..... 15
Figure 4 Lignin structure units and the linkage. Adapted from Kang et al. (2013). ..................... 17
Figure 5 Structural model of spruce lignin with 18 phenyl propane units. Adapted from
Freudenberg and Neish (1968).............................................................................................. 18
Figure 6 Main constituents of hemicellulose. Figure was adapted from Hansen and Plackett
(2008) (Hansen and Plackett, 2008)...................................................................................... 19
Figure 7 General structural formulas of arabino-(4-O-methyl-glucurono)-xylan. (XYL=xylose,
GLcA=methylglucuronic acid, ARA= arabinose). Figure was adapted from Hartman (2006)
(Hartman, 2006). ................................................................................................................... 20
Figure 8 General structural formulas of AcGGM. (GAL=galactos, GLC=glucose, MAN=mannos,
Ac=acetyl group). Figure was adapted from Hartman (2006) (Hartman, 2006). ................. 20
Figure 9 Theoretical reaction in which acetylated galactoglucomannan (AcGGM) is fully
benzylated exemplified by an acetylated mannose unit. Figure was adapted from Hartman
(2006) (Hartman, 2006). ....................................................................................................... 30
Figure 10 Theoretical styrene monomer attachment to AcGGM surface by plasma treatment or
vapor-phase grafting. Figure was adapted from Hartman (2006) (Hartman, 2006). ............ 31
Figure 11 Tensile stress-strain diagram ........................................................................................ 49
Figure 12 UV-Vis absorption spectrum (a) and transmittance spectrum (b) of bleached and
unbleached HPP hemicelluloses based films ........................................................................ 58
Figure 13 DSC thermograms of HPP hemicelluloses and their films with and without additives,
(a) bleached; (b) unbleached. ................................................................................................ 61
Figure 14 Tensile load-extension diagram ................................................................................... 64
Figure 15 Normal Q-Q Plot of tensile strength of hemicellulose films from bleached aspen
CTMP pulp fiber ................................................................................................................... 67
Figure 16 SEM morphology of HPP hemicellulose based films, a: bleached; b: unbleached..... 72

x

Figure 17 Relationship between films TS/ MOE and molecular weight/ distribution ................. 84
Figure 18 Calibration curve of dextrans with known molecular weights ..................................... 86
Figure 19 SEC elution profile of hemicellulose extracted from bleached CTMP aspen .............. 86
Figure 20 SEC expanded elution profile of hemicellulose from bleached CTMP aspen ............. 87
Figure 21 Relationship between the TS of hemicellulose films and hemicellulose DPs.............. 90
Figure 22 FTIR spectra of HPP hemicellulose samples, HPP, xylan and lignin (a) and its
expanded spectra (b) ............................................................................................................. 94
Figure 23 X-ray diffraction patterns of hemicelluloses and hemicellulose films, (a) bleached; (b)
unbleached (The signal of sorbitol is reduced by a factor of 2.5 times). .............................. 98
Figure 24 X-ray diffraction patterns of xylan films, from beechwood, birchwood and oat spelts.
............................................................................................................................................... 99
Figure 25 Xylan film from beechwood with different ratios of sorbitol and glutaraldehyde .... 119
Figure 26 xylan film from beechwood with different amount of additives ................................ 122
Figure 27 QQ Plots of TS, MOE and El from different treatments ............................................ 124
Figure 28 Size exclusion chromatography whole elution profiles from two columns system of
hemicelluloses from different approaches, (A: aspen; B: bleached; S: spruce; UB:
unbleached). ........................................................................................................................ 131
Figure 29 Size exclusion chromatography whole elution profiles from single column system of
hemicellulose from different approaches, (A: aspen; B: bleached; S: spruce; UB:
unbleached). ........................................................................................................................ 134
Figure 30 Size exclusion chromatography elution profiles of xylans, freshly made solution and
after 5 days storage, (a) xylan from birchwood; (b) xylan from oat spelts. ........................ 134
Figure 31 Schematic diagrams of MOE calculation, (a) stress-strain method; (b) tensile loadextension method. ............................................................................................................... 135

xi

1 BACKGROUND AND MOTIVATION
1.1 Introduction and research motivation
Multitudinous petroleum derived plastic materials are in widespread use in many areas
nowadays; about 34% of these are used as packaging materials (American Chemistry Council,
2014). This gives rise to increasing concern about the shortage of fossil fuels in the foreseeable
future, since fossil resources are considered non-renewable due to the long time period and
critical conditions required in their formation. Meanwhile, a strong concern about environmental
issues is caused by the wide utilization of non-degradable petroleum derived plastic materials
resulting in an intensive increasing of their disposal in landfills. Furthermore, the release of
greenhouse gases (e.g., CO2) when they are burned has attracted significant attention related to
global climate change.
An important approach to overcome the negative feedback from utilization of petroleum
derived plastics is to use degradable polymers from renewable resources as an alternative.
Biopolymers obtained from a variety of renewable resources have been investigated for
biodegradable packaging applications. The most well-known biopolymers are cellulose, starch,
soy protein, polylactic acid (PLA) and polyhydroxyalkanoates (PHAs), which have been
successfully commercialized (Table 1) (Martín-Closas and Pelacho, 2011).

1

Table 1 Examples of commercially utilized biopolymers. Adapted from Martín-Closas and
Pelacho (2011).
Biopolymer

Trademark

Manufacturer

Cellulose

NaturflexTM

Innovia Films (UK)

Biograde®

FKuR Kunststoff (Germany)

Mater-Bi®

Novamont (Italy)

Bioplast®

Biotec (France)

Soy protein

PRO-FAM®

ADM (Netherlands)

PLA

IngeoTM

NatureWorks LLC (USA)

Bio-Flex®

FKuR Kunststoff (Germany)

Goodfellow PHB

Goodfellow (UK)

Biomer®

Biomer (Germany)

Starch

PHAs

Wood, as an important renewable and biodegradable natural resource, has been used as a
fuel and as a construction material throughout history. The total timber volume in the US had
risen to 1013 billion cubic feet in 2007 (Smith et al., 2009). It represents great storage of a
renewable and biodegradable resource in the world. There are three main components in wood,
cellulose, hemicellulose and lignin; each of them represents an abundant renewable source of
biodegradable polymers. For the past few decades, cellulose as well as lignin has been the
subject of intensive research, development and commercialization, while hemicellulose, as the
second most abundant component in wood, is not well explored. For example, in the paper and
pulping industry, a substantial amount of hemicellulose is just treated as waste or simply burned
for energy.

2

Recently, the utilization of hemicellulose as a renewable biopolymer for biodegradable
packaging materials has become of great interest from the economic efficiency viewpoint. Many
studies have been carried out on the fabrication of hemicellulose based packaging materials with
sufficient mechanical and barrier properties. These research objectives are mostly focused on the
hemicellulose isolation, characterization, identification, and modification methods. Few studies
are available in the open literature dealing with the influence of different hemicellulose sources,
isolation methods and pretreatments on the final product properties such as tensile strength and
water vapor barrier, which are important properties for packaging applications.
The exploitation of hemicellulose as an alternative to petroleum based plastics for
packaging materials will earn more output value for wood industry, and therefore have a positive
impact on the utilization of wood resources and also the sustainability of packaging development.
Some people may refer to hemicelluloses in their studies, but I will use the singular form
―hemicellulose‖ in this study.

1.2 Objectives
The objectives of this dissertation are:
1) To set up an isolation method for extraction of hemicellulose from woody materials.
2) To fabricate hemicellulose based films using the extracted hemicelluloses.
3) To improve the properties of the hemicellulose based films for potential use in food
packaging.
4) To characterize the physical, mechanical and barrier properties of the modified
hemicellulose based films.
5) To study the influence of different hemicellulose sources and pretreatment on the
properties of hemicellulose based films.

3

2 LITERATURE REVIEW
2.1 Introduction
Plastics, as a specific category of polymer with many good characteristics such as light
weight, convenience of use and good barrier properties, have been dominant in the packaging
market for decades. At present, an overwhelming majority of plastic polymers are derived from
natural gas and petroleum, which are non-renewable fossil resources. They are either burned or
dumped into landfills after short time use, causing serious public environmental concerns with
respect to littering issues and the lack of landfill space because of their non-biodegradability
(Selke et al., 2004). Therefore, enhancing recycling/reusing of plastics or replacing plastics by
new biodegradable polymers have become the two strategies to solve these problems. For
developing new biodegradable polymers, biological resources attract the most attention due to
their renewability and biodegradability.
So far, many kinds of renewable resources have been investigated to produce
biodegradable polymers for replacement of petroleum based packaging polymers. However, it is
difficult to say that developing biodegradable polymers from renewable resources is the best
scenario for ―green‖ materials and processing techniques. Without a life-cycle environmental
impact analysis, the energy required and material consumed for treatment and various wastes
disposed during the production process may not be taken into consideration sufficiently.
Therefore, developing biodegradable polymers from renewable resources might not be as ―green‖
as it appears. However, from the point of view of landfill space saving and alternatives for nonrenewable fossil resources, it could still be an advantage. Furthermore, if the energy required for
developing biodegradable polymers from renewable resources comes from a ―new energy‖

4

source, like solar power, wind energy, water/tidal power, etc., this approach could still be more
sustainable based on the carbon dioxide balance point compared with petroleum based polymers
(Jiang and Zhang, 2013) .

2.2 Biodegradable polymers from renewable resources currently used in packaging
In general, three groups of renewable resources currently provide polymers for
production of biodegradable packaging materials. The first group are naturally occurring
biodegradable polymers, such as polysaccharides, proteins, and lipids, which can be directly
obtained from biomass (Plackett et al., 2010; Nussinovitch, 2012; Peelman et al., 2013; Kuorwel
et al., 2013; Félix et al., 2014; Shi and Dumont, 2014). The second group is synthesized using
monomers from renewable resources, such as polylactic acid (PLA). PLA is synthesized from its
monomer, lactic acid (LA), which can be obtained by bacterial fermentation of carbohydrates
like corn, sugarcane, potatoes, etc. (Gruber et al., 1992). The third group is synthesized by
certain microorganisms through fermentation, such as polyhydroxyalkanoates (PHAs). These are
the carbon and energy storage materials produced by certain bacteria in a situation of limited
nutrients but excess carbon sources (Byrom, 1994; Steinbüchel, 1991).

2.2.1 Cellulose
Cellulose is a natural homopolysaccharide belonging to the first group. It is the most
abundant biopolymer in the world with an annual production of over 75 billion tons (Habibi et al.,
2010). It can be found in various natural resources in the world, such as wood, abaca, bamboo,
jute, flax, hemp, etc. The purest natural cellulose comes from cotton, which is about 90%
cellulose. Cellulose molecules are composed of β-D-glucopyranose units linked together by β-(1-

5

4)-glucosidic bonds, forming a completely linear structure (Azizi Samir et al., 2005). Its
molecular chain has an average degree of polymerization (DP) of at least 9000-10,000. The
repeat unit of the molecular chain (Figure 1) consists of two glucose residues that contain three
free hydroxyl groups on each anhydroglucose ring.

Figure 1 Schematic representation of cellulose molecule. Figure was adapted from Azizi Samir et
al. (2005) (Azizi Samir et al., 2005).

Therefore cellulose molecules are peculiarly prone to form intra and intermolecular
hydrogen bonds. Such hydrogen bonds connect cellulose molecules to form microfibrils, which
possess both crystal regions (cellulose molecules are highly ordered) and amorphous regions
(cellulose molecules are less ordered). The microfibrils are bonded together in the same manner
to form fibrils and eventually cellulose fibers. Because of the supermolecular structure and
strong hydrogen bonds, cellulose has high tensile strength and is difficult to dissolve in most
solvents. But it also has a major disadvantage: it is very hydrophilic because of the presence of a
large number of hydroxyl groups in the molecular chain. This feature limits its utilization in
various applications, especially in the food packaging area, without modification.
Many efforts have been made to improve its properties for satisfaction of the
requirements for packaging materials. Cellulosic polymers are classified into three main groups:
regenerated cellulose (e.g. cellulosic fibers and cellophane), cellulose esters (inorganic cellulose
esters, e.g. cellulose nitrate, and organic cellulose esters, e.g. cellulose acetate) and cellulose
6

ethers (carboxymethyl cellulose). Cellulosic fibers have been used in packaging as paper and
paperboard for a long time in a broad range of food categories (Kirwan and Strawbridge, 2003).
Cellophane, a regenerated cellulose product, was patented in 1912 and found immediate
commercial use in packaging due to its good properties, such as low permeability to air, oils,
greases etc. It is obtained by dissolving cellulose in an alkali and carbon disulfide solution, then
reconverting to cellulose using a sulfuric acid/ sodium sulfate solution and finally softening
using glycerol (Carlisle, 2004). Nowadays, its products have broadly registered trademarks in
many countries. It can be laminated with paper, aluminum foil, PET and PE for food packaging
application (Kirwan 2003; Kirwan and Strawbridge 2003). A recent research hotspot is cellulose
derivatives, which can be obtained by esterification or etherification of the hydroxyl groups on
cellulose molecules. These cellulose derivatives can be used to produce thermoplastic materials
processable by injection molding or extrusion. Most of them display great film-forming
properties, but have not been exploited on an industrial scale (Cyras et al., 2009; Shen et al.,
2009). Another newly developed cellulose application involves generating nanoscale cellulosic
materials, like cellulose nanofibers (CNs), cellulose nanowhiskers (CNW) and cellulose
nanocrystals (CNC) or microfibrillated cellulose (MFC), which could be used as packaging film
substrates or additives resulting in good barrier and excellent mechanical properties (Pääkkö et
al., 2008; Plackett et al., 2010; Spence et al., 2010; Stevanic et al., 2012).

2.2.2 Starch
Starch, a major carbohydrate used for energy storage in higher plants, is another
substance belonging to the first group. It has two types of macromolecules in its native state in
plants: linear amylose (20%-25%) and branched amylopectin (75%-80%) (Galliard and Bowler,

7

1987; Jenkins et al., 1993; Gallant et al., 1997). Amylose has a similar molecular structure to
cellulose; it predominantly consists of (1→4)-linked D-glucopyranose units as its backbone but
with -glycosidic linkages; amylopectin is a highly branched macromolecule with the same
backbone units ( -D-glucopyranose units) but linked by (1→4) and (1→6) linkages and
possessing high molecular weight (Whistler and Daniel, 1984; Kainuma, 1984), as shown in
Figure 2 (Carvalho, 2013). These features make starch possess different stereochemical
characteristics, resulting in different physical and chemical behaviors from that of cellulose.

(a)

(b)

Figure 2 Amylose (a) and amylopectin (b) structures. Adapted from Carvalho (2013).
8

There are several crops commercially used for the production of starch. Maize, wheat,
and rice were the top three crops for starch production in 2005, with about 712, 631, and 622
million metric tons produced, respectively (FAO, 2007). Starch is not meltable and becomes
soluble in hot water; therefore unmodified starch films cannot meet the requirements for most
applications. To be useful as a food packaging material, it needs modification to adjust its
properties. Two strategies are commonly applied for starch use: 1) as a raw material for the
synthesis of other polymers; 2) directly as the target polymer macromolecule without changing
the molecular structure. The first strategy includes: i) using starch to produce monomers for
synthesis of other polymers, such as nonbiodegradable polyethylene or biodegradable polylactic
acid (Otey and Doane, 1984; Garlotta, 2001); ii) as a raw material for hydroxyl-containing
monomers like glucose or dextrin for the production of mixed compositions with other
monomers (Otey and Doane, 1984; Koutinas etal., 2004); iii) and as a raw material for
biopolymer production by fermentation using microorganisms (Kennedy et al., 1987; Robertson
et al., 2006). The second strategy is to modify starch by combination with other polymers,
esterification, etherification, grafting and reactive or melt extrusion of thermoplastic starch (TPS)
(Carvalho, 2013). Many types of polymers are blended with TPS to overcome its disadvantages,
such as moisture susceptibility, brittleness and low strength for packaging polymer or foam
applications (Fang and Hanna, 2001, 2002; Willett and Shrongen, 2002; Rodriguez-Gonzalez et
al., 2003; Huneault and Li, 2007; Parulekar and Mohanty, 2007). Starch can also be acetylated
for water resistance and used in packaging foams (Ganjyal et al., 2004). Commercial starchbased materials have been successfully used for packaging applications (Bastioli et al., 1991),
and are even tolerant to high water activity levels (Robertson, 2008; García et al., 2009).

9

However, edible and soluble packages for flavor components also effectively utilize starch in
packaging (Allen et al., 1963; USDA, 1971).

2.2.3 Protein
Besides cellulose and starch, protein is another abundant renewable biopolymer having
great potential for replacement of petroleum based polymers in food packaging applications.
These are heteropolymers that usually contain most of the 20 amino acids. Soy is the most
abundant source for this application with potential annual production of 60 billion pounds of soy
protein in US, followed by gluten from wheat (about 14 billion pounds), and zein from corn
(about 7.8 billion pounds) (Shukla and Cheryan, 2001). Besides plants, there are also proteins
from animals, such as whey and casein from milk for edible film applications (Mabesa et al.,
1980; Sensidoni and Peressini, 1997).

2.2.4 Other biopolymers
Animals can also supply biopolymers used as edible coating materials in food packaging,
such as gelatin, shellac and collagen (Robinson, 1953; Kuntz, 1964; Antoniewski and Barringer,
2010; Soradech et al., 2012) etc. In addition, polysaccharides such as pectin, carrageenan and
algin are types of carbohydrates obtained from plants and used for edible coating applications
(Kang et al., 2007; Baldwin, 1994; Keshri and Sanyal, 2009).
In addition, polylactic acid, belonging to the second group, is commercially available on
a large-scale for different applications. Its consumption in 2012 was about 187,600 tons per year,
according to the European Bioplastics Association. The most important market in volume for
packaging application is biodegradable packaging (Auras et al., 2004). It is mainly used as a food

10

packaging for fresh products, which usually have short shelf-life, such as vegetables and fruit
(Domenek et al., 2011).
Poly-3-hydroxybutyrate (PHB), the polyester produced by certain bacteria processing
glucose, is an important member of the polyhydroxyalkanoate (PHA) family. Large-scale
commercial production of PHAs uses fermentation technologies. Homopolymer PHB is a brittle,
crystalline thermoplastic material and it undergoes thermal decomposition just at its melting
point. Therefore, extensive efforts have been made to overcome its drawbacks by synthesis of
copolymers for biodegradable applications (Rudnik, 2013 a).

2.3 Hemicellulose
Although modified biopolymers such as cellulose derivatives, thermoplastic starches or
soy protein based films have been successfully used in packaging, renewable biopolymers are
still currently of central interest due to the potential of replacement of conventional petroleum
based polymers typically used in food packaging (De Vlieger 2003; Sommer and Kunz, 2012).
Hemicelluloses recently have gained increasing interest in this type of application, not only due
to their biorenewability and biodegradability, but also their lower cost and avoidance of
competition with food supplies. In the past, the application of hemicellulose involved their
conversion into sugars, chemicals and fuel or directly as sources of heat energy (Dumitriu, 1996).
Even though other uses of hemicellulose in high-technology products have been investigated and
developed, the commercial utilization of hemicellulose is still limited. Thus, adding extra value
by using the hemicellulose as an alternative to petroleum based packaging materials would
benefit the development of sustainable packaging and reduction of environmental issues.

11

2.3.1 Hemicellulose sources
As the second most abundant family of polymers after cellulose, estimated global annual
production of hemicellulose is around 60 billion tons. It can be obtained from various sources,
not only from wood, but also from bamboo, grass, banana stems, agricultural residues, etc. It
broadly exists in the plant kingdom; you can even find it in the ocean and other aquatic systems
(Sun et al., 2004). For a large scale supply, in addition to the herbaceous plants that are attracting
more and more interest from researchers due to the large amount of agricultural residues, wood
plays an important role as a source of high quality hemicellulose. Not only sawdust from the
wood industry, but the side streams during production of thermomechanical pulp could be
sources for hemicellulose for value-added products or byproducts. This is called value prior to
pulping (VPP) in the paper and pulping industry. Additionally, paper grade pulp can be another
large scale source of hemicellulose.

2.3.1.1 Wood sources

2.3.1.1.1 The state of wood resources
Wood, as a precious renewable resource, has supplied a huge amount of raw materials since
ancient times before recorded history. It is the most important biological and renewable resource
in the world. For a long time, the utilization of wood remained at the macro level, such as use as
a fuel and as a raw material for construction and furniture. The growth of the wood pulp and
paper manufacturing industry led to use of wood as fiber. With increasing interest in bio-based
materials, there is interest in utilizing wood at the micro level. New technologies have been
developed to convert various components of wood into high value products, such as lignin glue,
biodegradable food packaging, alternative agents for rubber tires, anti-bacterial medical agents,
12

and high strength fabrics or nanocomposites. Considering the benefits of wood to society as well
as the interest in sustainable development, it is critical to utilize wood more effectively and
efficiently.
Wood is a hard and fibrous structural tissue consisting of three main components: cellulose
(40%-45%), hemicellulose (20%-30%) and lignin (20%-25). Besides these three main
components, wood also contains other minor components, ash and extractives, which include
wax, pectin, resin acids, etc. (Sjöström, 1993). Since most research efforts have focused on
cellulose and lignin, these two components of wood have been extensively explored and utilized.
Cellulose based products are broadly used. As mentioned before, cellulose fibers in the form
of paper and paperboard, from one of the largest chemical industries in the world, the pulp and
paper industry (Wise and Jahn, 1952), have been used in packaging for a long time in a wide
range of food categories including fresh, dry, frozen, and liquid foods or beverages (Kirwan and
Strawbridge 2003). Besides packaging applications, cellulose is also used in liquid crystal
displays (Sohn et al., 2003), hollow fibers for artificial kidney dialysis (Sargent, 2007), and
components or additives in pharmaceutical and food applications (Felton and Porter, 2013;
Mikkelsen et al., 2011), among others. Recently, TEMPO-oxidized cellulose nanofibers (TOCN)
were investigated for potential application in ion-exchange materials for water and highperformance air-filters (Isogai et al., 2011).
Lignin accounts for only 30% of the biomass weight but it contributes 40% of the fuel value,
which makes it a strong candidate for biofuel products (Holladay et al., 2007). Based on its
phenolic functionalities and properties, lignin attracts more attention in higher value-added
renewable product applications. Lignin is a good source for valuable chemicals (aromatic
aldehydes and phenolic products) after it is broken into smaller molecular units because of its

13

sustainable, abundant, phenolic and carbon-neutral characteristics. A series of approaches have
been applied to generate biofuels and valuable compounds from lignin, including enzymatic
degradation (Xia et al., 2003), photo-catalyzed degradation (Kansal et al., 2008), ionic liquids
degradation (Binder et al., 2009), electrochemical degradation (Tian et al., 2010), microwave
irradiation oxidation (Ouyang et al., 2010), hydrothermal conversion and pyrolysis (Wang et al.,
2009; Jiang et al., 2010a, b; Mu Wei et al., 2013). Among these approaches, hydrothermal
conversion of lignin has been extensively investigated (Kang et al., 2013).

2.3.1.1.2 Internal environment of hemicellulose in wood
In wood cell walls, the three main components do not exist in isolation. Cellulose fibrils
are embedded in hemicellulose and lignin matrix. The basic structure/composition analysis and
the application of these wood components as raw materials require as far as possible to isolate
them completely from wood with as little as possible damage or change to the component
structure or molecular weight. The extractions of these components all involve the breakage of
the connection among these components. It is necessary to fully understand the properties of the
three main components (e.g., composition, structure, physical and chemical interactions) in order
to isolate high quality components.

2.3.1.1.2.1 Cellulose
As mentioned earlier in this chapter, cellulose, as the primary component in the wood cell
wall, accounts for approximately 40-45% of the dry weight of wood in most species. It is a linear
polymer with a rigid backbone that crystallizes and can be loadbearing (Sjöström, 1993).

14

Cellulose is a hydroxyl group-rich homopolymer with 3-dimensional supermolecular structure.
Figure 3 shows the molecular and supermolecular structure of cellulose from wood (Dodson,
2012).
Growth Ring
mm
Tree

Transverse Section

Cellular Structure
500μm

m

cm

Figure 3 Cellulose structures in trees from logs to molecules. Adapted from Dodson (2012).

2.3.1.1.2.2 Lignin
Lignin is a carbon-rich heterogeneous biopolymer containing aromatic structural units. It
is the only renewable source of aromatics, and represents an important constituent of biomass.
The estimated potential lignin production is more than 50 million tons per year from the pulp and
paper industry (Gosselink et al., 2004). Lignin has three basic structural units: p-coumaryl
alcohol, coniferyl alcohol and sinapyl alcohol. Hardwood lignin is primarily derived from
sinapyl alcohol and coniferyl alcohol. Softwood lignin is mainly derived from coniferyl alcohol.
15

It is a natural polyphenol with a 3 dimensional network structure constituted of the three basic
structural units linked by C-C and ether bonds (Hüttermann et al., 2001), as shown in Figure 4
(Kang et al., 2013).
No effective analytical method has yet been employed to characterize the natural
structure of lignin. Its structural model can be deduced by biosynthesis or chemical degradation
combined with NMR and FTIR or Raman spectra analysis. Figure 5 shows a structural model of
spruce lignin with 18 phenyl propane units (Freudenberg and Neish, 1968).

16

Figure 4 Lignin structure units and the linkage. Adapted from Kang et al. (2013).

17

Figure 5 Structural model of spruce lignin with 18 phenyl propane units. Adapted from
Freudenberg and Neish (1968).

18

2.3.1.1.2.3 Hemicellulose
Hemicellulose from wood is a heteropolysaccharide with a series of hexose and pentose
units that are linked in an irregular manner, in contrast to the regular arrangement in cellulose. It
is entirely amorphous with no crystal structure in its natural state. It has a low degree of
polymerization compared with cellulose, usually between 100 and 200 (Rowell, 2012). It mainly
contains the following sugar residues: D-glucopyranose, D-mannopyranose, D-galactopyranose,
L-arabinofuranose, D-xylopyranose, D-glucopyranosyluronic acid and D-galactopyranosyluronic
acid with minor amounts of other sugars (Timell, 1967; Sun et al., 2004; Spiridon and Popa,
2008), as shown in Figure 6 (Hansen and Plackett, 2008). The high variability of hemicellulose
composition and structure makes it very challenging to characterize.

Figure 6 Main constituents of hemicellulose. Figure was adapted from Hansen and Plackett
(2008) (Hansen and Plackett, 2008).

The general composition of hemicellulose varies between hardwoods and softwoods.
Figure 7 shows the main hemicellulose in hardwood, O-acetyl-(4-O-methylglucurono) xylan
(Hartman, 2006), while the predominant hemicellulose type in softwood is O-acetylgalactoglucomannan (AcGGM), as shown in Figure 8 (Hartman, 2006).
19

Figure 7 General structural formulas of arabino-(4-O-methyl-glucurono)-xylan. (XYL=xylose,
GLcA=methylglucuronic acid, ARA= arabinose). Figure was adapted from Hartman (2006)
(Hartman, 2006).

Figure 8 General structural formulas of AcGGM. (GAL=galactos, GLC=glucose, MAN=mannos,
Ac=acetyl group). Figure was adapted from Hartman (2006) (Hartman, 2006).

As the compositions and structures of hemicelluloses above show, they, like cellulose,
are rich in hydroxyl groups, but they also can possess acetyl groups.
As a result of these functional groups and their macromolecular structures, the three main
components of wood are connected to each other tightly and systematically by physical binding

20

and chemical interaction. Between cellulose and hemicellulose, hydrogen bonding and molecular
chain intermixing are the main forms of binding (Scheller and Ulvskov, 2010); while covalent
bonds (mainly R-benzyl ether linkages) are the primary connections between hemicellulose and
lignin (Deniaud et al., 2003).
2.3.1.2 Pulp source
In the modern pulping industry, the steam liquid after pulping and also the pulp itself are
important sources for hemicellulose due to the large volume available in the world. Total
production of wood pulp (air dried pulp) for paper and paperboard in the US was reported to be
50301×103 metric tons in 2012; market wood pulp (air dried pulp) was reported as 8392×103
metric tons (FAO, 2012-2017). Pulp has become an abundant raw material worldwide.
Several pulping processes are currently used in today’s industry: mechanical pulp (RMP),
thermomechanical pulp (TMP), chemithermomechanical pulp (CTMP), chemical pulp (mostly
Kraft pulp), recycled pulp, organosolv pulp and other alternative pulping methods. The
chemithermomechanical pulp process is to pretreat wood chips with sodium carbonate, sodium
hydroxide, sodium sulfite and other chemicals before refining them with refiner plates which are
similar to mechanical pulp equipment. The conditions during the chemical treatment are much
milder than in a chemical pulping process, such as lower temperature, shorter time, and less
extreme pH. So the fiber will have less degradation and the lignin will not be removed as much
as in a chemical pulping process. Organosolv pulping uses organic solvents, such as methanol,
ethanol, formic acid and acetic acid often in combination with water at temperatures above
140 °C to break down lignin and hemicellulose into soluble fragments in the solvent liquor. It
nearly removes all the hemicellulose in the pulp. Therefore, organosolv pulp cannot be a
hemicellulose source. Overall, chemical pulping predominates in world pulp production (more

21

than 80%), and bleached Kraft pulp accounted for more than half of chemical pulp production in
2012 (FAO, 2012).
2.3.2 Hemicellulose definition and utilization
The term hemicellulose was originally proposed to designate polysaccharides that can be
extracted from higher plants by alkaline solution (Schulze, 1891). Hemicellulose can be divided
into four classes of polysaccharides with different structures, based on the current stage of
knowledge (Ebringerová et al., 2005): (a) xylans, (b) mannans, (c) β-glucans with mixed
linkages, and (d) xyloglucans.
Based on the primary structure of various plant xylans, xylan-type polysaccharides can be
divided into two types, homoxylans and heteroxylans. Homoxylans are found in seaweeds and
green algae, and have a backbone consisting of xylopyranose residues linked by (1→3) or mixed
(1→3, 1→4) glycosidic linkages (Ebringerováand Heinze, 2000; Yamagaki et al., 1997). Xylans
from terrestrial plants are heteroxylans with a β-(1-4)-D-xylopyranose backbone with branches
comprised of short carbohydrate chains. These side chain groups include L-arabinose, Dglucuronic acid or its 4-O-methyl ether, and/or various oligosaccharides. D-glucose, L-arabinose,
D- or L-galactose and D-xylose are commonly the main constituents of these oligosaccharides.
This type of xylans include glucuronoxylans, (arabino)glucuronoxylans, arabinoxylans,
(glucurono)arabinoxylans and other complex heteroxylans (Ebringerováet al., 2005).
In the past, utilization of hemicellulose was mainly focused on converting it to sugars and
chemicals, such as sweetening agents, thickeners, and emulsifiers used in the food industry
(Dumitriu, 1996). One industrial scale use of hemicellulose is to convert xylan to xylose, xylitol
and furfural. Xylitol has been applied in a variety of food products (Ebringerová and Heinze,
2000). In the pulp and paper industry, retaining the hemicellulose in the pulp has been shown to

22

improve both the mechanical properties of the paper and the yield (Schönberg et al., 2001; Ban et
al., 2011; Oksanen et al., 2011). Recently, more attention has been placed on the application of
hemicellulose as a raw material for high-technological products (Hansen and Plackett, 2008)
such as cationic polymers (Ebringerova et al., 1994), hydrogels (Gabrielii et al., 2000), longchain ester derivatives (Sun et al., 1999; Gabrielii et al., 2000; Petzold-Welcke et al., 2014), and
thermoplastic hemicellulose derivatives (Sun et al., 1999; Jain et al., 2000). Meanwhile,
hemicelluloses used as ingredients in the formulation of flexible packaging films attract more
and more attention from researchers. The definition of flexible packaging is a package or
container made of flexible or easily yielding materials that, when filled or closed, can be readily
changed in shape. The construction may be of paper, plastic film, foil or any combination of
these, including rollstock, bags, pouches, labels/wraps, lidding, shrink sleeves and stretch film
(Schottland, 2004). The flexible packaging market in U.S. is about $20 billion per year. Film is
defined as an optional term for sheeting having a nominal thickness no greater than 0.25 mm
(0.010 in, ASTM D883).

2.4 Problems for exploitation of hemicellulose in flexible packaging films
Applicable food packaging materials must have adequate mechanical properties and
provide a sufficient barrier to oxygen, water vapor, light, bacteria and/or other contaminants in
order to prevent food deterioration to provide a shelf life that is sufficiently long. So for
hemicellulose to be used as a packaging material, feasible and reliable hemicellulose extraction
and modification techniques are essential because of the inherent characteristics of hemicellulose.

23

1) Hemicellulose has a complicated composition and structure in contrast to cellulose; it
is a series of hexose and pentose units that are linked in an irregular manner. Different plant
species, plant locations, growth stages and isolation processes can result in significant variation
in hemicellulose composition and structure (Schröder et al., 2001; Sun et al., 2005; Lai et al.,
2006). The high variety of hemicellulose composition and structure from different sources and
processes make it very challenging to process hemicellulose in a competitive and guaranteed
industrial scale; hence hindering the extension of the utilization of hemicellulose.
2) In general, hemicellulose from wood has a much lower degree of polymerization than
cellulose, usually between 100 and 200 (Rowell, 2012). It is branched, and difficult or
impossible to crystallize. Thus, hemicellulose comparatively has lower strength than cellulose.
3) Hemicellulose has an abundance of hydroxyl groups in its structure, resulting in strong
hygroscopicity. This makes hemicellulose based films have a strong tendency to absorb moisture
from the air, resulting in not only low water vapor barrier but also negatively impacting its
physical and mechanical properties. Additionally, the moisture content promotes the growth of
microorganisms, and reduces the protection for the food.
4) The inherent nature of hemicellulose and its film forming procedure also mean
hemicellulose based films have an oxygen transmission rate that needs to be improved for
applications with higher requirements.
5) Hemicellulose is a nutrient for microorganisms. Anti-bacterial and anti-mold agents
are therefore necessary for some specific food packaging applications.
Thus it can be said that in order to transfer the potential of hemicellulose as a packaging
material into reality, the characteristics of hemicelluloses from different sources and
pretreatments should be clarified in order to clearly understand the influences of these

24

characteristics on their final performance. In addition, hemicellulose based packaging films
should also possess sufficient mechanical/physical properties, adequate oxygen/moisture barrier
properties and anti-bacteria/anti-mold properties (especially for some special applications). This
needs suitable, reliable and cost efficient isolation methods based on a sound understanding of
the influences from the different sources and pretreatments to obtain hemicellulose with as high
quality as possible for fabrication of packaging materials. Then corresponding measures could be
taken to modify the hemicellulose in order to obtain enhanced properties that will satisfy the
requirements for packaging materials.

2.5 Review of studies on hemicellulose for packaging applications
2.5.1 Different sources for hemicellulose and their compositions and structures
Research shows that the composition and structure of hemicellulose varies between
different plant species, plant locations, growth stages, and also the different isolation processes.
Significant variations of hemicellulose composition and structure could be obtained and thus
produce markedly different hemicellulosic products. (Gatenholm and Tenkanen, 2004; Lindblad
and Albertsson, 2005; Sun et al., 2005; Lai et al., 2006).

2.5.2 Isolation methods
Hemicellulose in the plant cell wall is bound to cellulose and lignin. The interaction
among these three components and also the relatively large molecular size make it difficult to
diffuse out of the matrix during hemicellulose extraction. Several most commonly used
extraction methods are described below.

25

2.5.2.1 Steam treatment
When wood is exposed to high temperature in a moisture-rich environment, the ester
bonds in the hemicellulose will be cleaved and hence produce some organic acids, mainly acetic
acid. The release of these organic acids further catalyzes the auto-hydrolysis of the glycosidic
bonds in the hemicellulose and other wood polymers. Therefore, a partially depolymerized low
molecular weight hemicellulose in aqueous solution is obtained. Due to the degradation of the
hemicellulose, this trealtment results in a low quality hemicellulose product with low yields and
contamination of cellulose and lignin. O-acetyl groups on the hemicellulose molecular chain can
be retained. A yield of 8% (wt.) of hemicellulose was reported after spruce steam treatment for 2
min at 200°C (Palm and Zacchi, 2003).

2.5.2.2 Microwave treatment
This technique employs microwaves to irradiate wood for heating in water. The
mechanism is the same as in the technique above, and bears the same disadvantage, except for
the uniform heating in this treatment. Meanwhile, new problems like long time requirements for
heating and cooling result (Teleman et al., 2000; Lundqvist et al., 2002).

2.5.2.3 Explosion treatment
Explosion treatment is also a hydrothermal treatment like steam treatment. Hence it leads
to a similar release of acids which can promote auto-hydrolysis in the wood polymers together
with the other disadvantages of steam treatment. The main difference between steam treatment

26

and steam explosion is the rapid depressurization that takes place at the end of steam explosion.
This could break the wood materials to small size fibers or fiber bundles, increasing the exposure
surface of wood materials to the cooking liquor (Ahvazi et al., 2007; Martín-Sampedro et al.,
2011a, b). Meanwhile, it could bring more opportunities for contamination from cellulose and
lignin. In other words, the parameters during the steam explosion need to be designed and
controlled appropriately in order to obtain better quality hemicellulose (Wu et al., 1999;
Krawczyk et al., 2008;).
So generally speaking, these hydrothermal techniques are not suitable for extraction of
high quality hemicellulose owing to the disadvantages mentioned above. They are more likely to
be used as a treatment to remove hemicellulose, obtaining the cellulose for other uses like in
Kraft pulping (Martín-Sampedro et al., 2014).

2.5.2.4 Hemicelluloses from process water
In the wood industry, there is a common hydrothermal treatment step to soften the lumber
before processing (Edlund et al., 2010). And in the paper and pulping industry, the extracted
streams from CTMP mills are rich in hemicellulose (Saadatmand et al., 2012; Liu et al., 2012;
Mikkonen and Tenkanen, 2012). Hemicelluloses can be found in various process liquors
(Persson et al., 2005). Microfiltration (Krawczyk and Jonsson, 2011), ultrafiltration and
diafiltration (Persson et al., 2007; Krawczyk et al., 2008) are usually the techniques that are
employed to separate hemicellulose from the other substances in the process water. Thus,
hemicellulose as a by-product could add extra value and facilitate reuse of the waste to reduce
cost.

27

An economic evaluation of isolation of hemicelluloses from process streams has been
done for thermomechanical pulping of spruce (Tobias Persson et al., 2007). The report stated that
it is possible to produce hemicellulose based films with good oxygen barrier properties at a price
that is competitive with the materials used today.

2.5.2.5 Alkaline extraction
Alkaline extraction has been proved to be a promising process for complete utilization of
lignocelluloses without environment issues (Peng et al., 2009). This technique dissolves
hemicellulose by cleavage of the R-ether linkages between the lignin and hemicellulose and the
ester bonds between lignin and/or hemicelluloses while at the same time swelling the cellulose to
promote the release of hemicellulose from the matrix (Jackson, 1977; Spencer and Akin, 1980).
Hence, the hemicellulose extracted by this technique could be of high quality without severe
degradation of hemicellulose or critical contamination with cellulose or lignin (Sun et al., 2001;
Peng et al., 2013; Sun et al., 2013). Through this process, alkali-soluble hemicelluloses can be
easily utilized for valuable products (Sun and Hughes, 1998; Gabrielii et al., 2000; Grøndahl et
al., 2004; Ruzene et al., 2008; Peng et al., 2009; Agneta et al., 2009; Zhang et al., 2011; Laine et
al., 2013; Alekhina et al., 2014).

2.5.3 Influence of composition and structure on the final product properties
Hemicellulose is a type of heteropolymer, with various compositions and structures. This
nature makes it capable of having enormously different properties when involving interaction
with other materials. The different constituents, substitution, molar mass and structures of
hemicellulose can lead to diverse reactivity. Many studies have examined the influence of
28

hemicellulose composition and structure (substituent patterns, molar mass and crystallinity) on
the final film properties, which is well correlated with the substitution ratio, for example Ara/Xyl
ratios of arabinoxylan (Teleman et al., 2000; H ̈ ije et al., 2008; Zhang et al., 2011; Stepan et al.,
2012).

2.5.4 Modification methods
A number of different physical and/or chemical approaches have been investigated for
fabrication of flexible hemicellulose based films with enhanced mechanical and barrier
properties for packaging applications, as discussed below:

2.5.4.1 Addition of plasticizer and cross-linking agents
This technique commonly employs sorbitol, glycerol, propylene glycol or xylitol as a
plasticizer to make the hemicellulose based film more flexible to resolve its brittleness. The
studies showed that hemicellulose film with sorbitol has better properties overall; at about 20%
(wt.) addition based on the dry weight, the film had the highest tensile strength, and at the same
time greater elongation compared to the other plasticizers. Cross-linking agents, such as glyoxal,
glutaraldehyde and genipin, are usually used to enhance the strength of the hemicellulose based
films. They can build covalent bonds within and/or between the hemicellulose molecular chains.
Thus they can remarkably improve the tensile strength and potentially contribute to the oxygen
and moisture barrier properties (Gabrieliia et al., 2000; Zhang and Whistler, 2004；Grøndahl et
al., 2004; Mikkonen et al., 2009; Hansen et al., 2012).

29

2.5.4.2 Esterification and etherification
The large amount of hydroxyl groups on the hemicellulose molecular chain provides
excellent opportunity for property changes by substitution reactions. Changing functional groups
can result in regulation of hemicellulose solubility, stability, and film-forming ability in order to
obtain the desired properties. The techniques make the hydroxyl groups of hemicellulose react
under certain conditions to be esterified or etherified to obtain hydrophobic carboxymethyl
hemicellulose and benzyl hemicellulose, thus improving the moisture barrier (Hartman et al.,
2006; Kochumalayil et al., 2013). The theoretical reaction of benzylation of an acetylated
mannose unit is shown in Figure 9 (Hartman, 2006).

Figure 9 Theoretical reaction in which acetylated galactoglucomannan (AcGGM) is fully
benzylated exemplified by an acetylated mannose unit. Figure was adapted from Hartman (2006)
(Hartman, 2006).

30

2.5.4.3 Surface grafting
Surface grafting usually employs plasma to treat the hemicellulose film surface,
producing functional groups for the next stage. The gas plasma could be oxygen, argon, nitrogen
or air. After plasma treatment, the film surface is then exposed to hydrophobic materials like
styrene resulting in vapor-phase grafting with styrene to make the film surface linked with
hydrophobic styrene molecules, as shown in Figure 10 (Hartman, 2006). This technique allows
fabrication of hemicellulose films with hydrophobic surfaces to reduce moisture sensitivity
(Hartman et al., 2006; Kochumalayil et al., 2013).

Figure 10 Theoretical styrene monomer attachment to AcGGM surface by plasma treatment or
vapor-phase grafting. Figure was adapted from Hartman (2006) (Hartman, 2006).

2.5.4.4 Lamination
This method allows hemicellulose to laminate with a hydrophobic material. The
hydrophobic material mostly determines the properties of the final products. By lamination of an
AcGGM alginate blend film with benzylated galactoglucomannan (GGM), excellent barrier
properties were obtained for water vapor with 8 cm3

31

m-2 (24 h-1) kPa-1 at 83% RH (Hartman

et al., 2006).

2.5.4.5 Surface coating
This technique is aimed at improving barrier properties. For obtaining a good gas and
vapor barrier layer, the coating material needs to be pore-free and it is better if it is hydrophobic.
Al2O3 can be deposited on the surface of polymers by the atomic layer deposition (ALD)
technique, forming a high-quality pore-free barrier film. Al2O3 coated galactoclugomannan
(GGM)-coated board (GGM appr. 9g/m2) reduced the OTR value to close to the oxygen barrier
level required for dry food applications (Hirvikorpi et al., 2011). TEMPO-oxidized cellulose
nanofibrils possess free carboxyl groups, which could change the cellulose nanofibrils to
hydrophobic, and thus remarkably increase the barrier properties (Isogai et al., 2011).
2.5.4.6 Composites
Hemicellulose blended with nanoscale cellulose can remarkably improve the mechanical
properties, such as stiffness, strength and stretch. Nearly 1.5 times the stiffness (7.3 GPa) and
improved strength (143 MPa) compared to pure hemicellulose film were observed in a high
molecular weight arabinoxylan based hemicellulose film, and the stress at break (7.2%) was
nearly twice that of the pure hemicellulose based film. Nanoclay blending with hemicellulose
reduced the water vapor transmission rate (WVTR) of the composite by about 5%. However, in
combination with crosslinking agents, WVTR was reduced by about 10%. Even though the
nanoscale substance addition in hemicellulose reduced the water/water vapor permeability, it was
still two orders of magnitude larger than in current commercially used packaging films (Edlund
et al., 2010; Doroudgarian, 2011; Stevanic et al., 2012; S rossy et al., 2012；Saadatmand et al.,
32

2012).

2.5.4.7 Enzymatic modification
The properties of polymers are related to the degree of polymerization as well as to
substitution. The use of enzymes can change the amounts of the side groups of hemicellulose,
hence changing both its solubility and crystallinity. Removal of side groups can result in a closer
association of hemicellulose molecular chains, and thus contribute to crystallization and
influencing the performances of the hemicellulose products (Settineri and Marchessault, 1965;
Chanzy et al., 1979).

2.6 Economic evaluation for hemicellulose application
To turn the potential use of hemicellulose as a packaging material for relieving the
dependence of society on petrochemicals into reality, the isolation method of hemicellulose
should be cost-efficient. As a promising hemicellulose source, the process streams in
thermomechanical pulp mills have attracted much attention for hemicellulose extraction. Persson
et al. in 2007 performed an economic evaluation of a hemicellulose isolation method from the
process streams in thermomechanical pulp mills by membrane filtration. They proposed that
using hemicellulose as an oxygen barrier could be possible at a price that is competitive with the
materials used today. Nevertheless, the economic evaluation of hemicellulose isolation from
other sources (wood, agricultural residues, grass) is hardly seen in the open literature.

33

3 MATERIALS AND METHODS
3.1 Introduction
Hemicellulose, as the second most abundant renewable biopolymer on earth, has attracted
a great deal of attention for its potential in manufacturing of high-value-added products, like
food packaging materials. For these materials, adequate mechanical properties (strength and
flexibility), sufficient barrier to oxygen and water vapor, and in some cases, preventing light,
bacteria and/or other contaminants are necessary in order to keep food from deterioration to
provide a sufficiently long shelf life. Therefore, efficient measures are necessary to take in order
to overcome the drawbacks of hemicellulose to provide adequate properties.
Many studies have been carried out to fabricate reliable packaging materials with
sufficient mechanical and barrier properties from hemicellulose. Research objectives mostly
focus on the isolation, characterization, identification, and modification methods. Few studies are
available in the open literature dealing with the influence of different hemicellulose sources,
isolation methods and pretreatments on the properties important for packaging materials such as
tensile properties and water vapor transmission.
Efforts need to be made to fully understand the influence on the properties of
hemicellulose for selection and/or adjustment of isolation methods to provide the properties
needed for large industrial scale applications. And also corresponding measures could be taken to
modify hemicellulose products in order to obtain enhanced properties, to satisfy the requirements
as packaging materials.
Wood is extensively used for pulping to make paper and paperboard. In USA, the forest
area slightly increased from 296,335 (1000 ha) in 1990 to 304,022 (1000 ha) in 2010 (FRA,
2010). Aspen (Populus tremula) is a very commonly seen hardwood species. Spruce is

34

commonly found in the northern temperate and boreal (taiga) regions of the earth. They are both
important species for the pulp and paper industry. Chemi-thermo-mechanical pulp (CTMP)
aspen hardwood and spruce softwood can offer ideal raw materials for the isolation of
hemicellulose with high yield and reasonably high molar mass, due to their high production, and
low content of lignin and other contaminants (Fuhrmann and Krogerus, 2009). Hybrid poplar, as
one of the hemicellulose sources in this study, is one of the fastest growing wood species. It can
be harvested from December to June. It grows broadly in temperate and cold temperate zones in
the northern hemisphere. Like aspen, hybrid poplar is a popular species in many countries,
including the United States. In addition, utilization of wood, unlike cereal plants such as rye,
wheat, oat, corn, etc., does not compete with food for people or livestock. So, aspen CTMP
(CTMP A), hybrid poplar powder (HPP) and spruce CTMP (CTMP S) are the three types of raw
materials used as hemicellulose sources in this study.
Hemicellulose has many hydroxyl groups in its molecular chain. These groups make
hemicellulose very hydrophilic. But also they provide the possibility to be modified by cross
linking, changing them to ester bonds, to increase the intermolecular interaction/connection for
improvement of the mechanical properties. Meanwhile, it would be an effective approach to
improve their hydrophobic properties. For resolving the brittleness of hemicellulose based film,
plasticizer is needed to increase its flexibility. Sorbitol is proved to be a promising plasticizer for
arabinoxylan based film (Hansen et al., 2012). Glutaraldehyde has been used as a cross-linking
agent in many applications, such as in bioprostheses, the fixation of cells, the immobilization of
enzymes, and the delivery of controlled drug (Jayakrishnan and Jameela, 1996). It is also
allowed to be used as a component of paper and paperboard in contact with aqueous and fatty
food (21CFR176.170). EPA (Environmental Protection Agency) in 2007 published the

35

reregistration eligibility decision for glutaraldehyde (EPA 739-R-07-006). It is used in food
handling and food storage establishments as an antimicrobial agent. Glutaraldehyde-containing
products are also approved for use in aquatic areas. EPA concluded that the special hazard-based
FQPA Safety Factor for glutaraldehyde should be removed.
Thus, in this study, sorbitol and glutaraldehyde were employed as additives to improve
the properties of hemicellulose based films for potentially use for packaging application.
Commercial xylans (Sigma-Aldrich) as a perfect model for hemicellulose were
investigated in advance to figure out the proper additive amounts and their ratio in the xylan
based films. The degree of polymerization (DP) of these xylans and tensile properties of their
films were tested as well, for clarification of the influence of DP on the properties. Then,
hemicelluloses from different sources (hardwood & softwood; pulp & powder) and pretreatments
(chemithermomechanical treatment & mechanically ground; bleached & unbleached) were used
in this study to fabricate hemicellulose based films with sorbitol and glutaraldehyde as the
plasticizer and cross-linking agents, respectively.
The alkaline extraction method was selected in this study for isolation of hemicellulose,
because the structure of alkaline extracted hemicellulose is quite similar to its native status
except for the removal of O-acetyl groups under alkaline conditions compared with the other
methods (Teleman et al., 2000). The equipment used in this method is not required to support
high temperature and pressure, which makes it very easy to conduct in the laboratory.
Furthermore, the chemicals used in this method are relatively easy to handle.

36

3. 2 Materials and chemicals
Xylans from beechwood, birchwood and oat spelts purchased from Sigma-Aldrich were
used as the hemicellulose models in this study. Hybrid poplar was ground to reduce its size to
pass mesh #35 in order to increase its specific surface area exposure under the reaction
conditions, which helps hemicellulose release from its matrix in the wood cell wall. Thus, three
types of raw materials, hybrid poplar powder (HPP), aspen CTMP (CTMP A) and spruce CTMP
(CTMP S), were used for alkaline extraction of hemicellulose. The raw materials were either
bleached or unbleached before alkaline extraction. Hence, hemicellulose was prepared through
six

different

approaches.

Detailed

preparation

procedures

are

explained

below.

Ethylenediaminetetraacetic acid (EDTA, Sigma-Aldrich), ethanol (190 proof, Baker), and filter
paper (Waters) were the chemicals and reagents used in the hemicellulose preparation steps.
HPLC water (Baker), dextran with different molecular weights (Sigma-Aldrich), sodium nitrate
(Sigma-Aldrich) and sodium hydroxide (Sigma-Aldrich) were the chemicals used in the
hemicellulose analysis procedures. Sorbitol and glutaraldehyde (50%) were purchased from
Sigma-Aldrich and used as the film additives. Disodium 2, 2’-bicinchoninate, sodium carbonate,
sodium bicarbonate, copper sulfate pentahydrate and L-serine were purchased from SigmaAldrich and used for DP measurement by the reducing-end method.
3.3 Fabrication of xylan based films by solvent casting
3.3.1 Function of additives
Xylan from beechwood (0.85g, dry weight) was dissolved in 90 g DI water by heating at
69±1℃ for about 30 min with magnetic stirring. After cooling down to room temperature, a total
of 0.25 g sorbitol and glutaraldyhyde (GA) were added to this xylan solution at ratios of 3:1, 2:1,
1:0, 0:1, 1:1, 1:2, 1:3 sorbitol:GA. Each of these solutions was then magnetic stirred again and

37

transferred evenly to 3 petri dishes (100 mm diameter × 15 mm height) by passing through a
#230 mesh screen to remove particles. These samples were then evaporated at room temperature
(23℃) for about 3 days to form films. Then they were transferred to a room maintained at 23℃
temperature and 50% relative humidity and conditioned for 2 weeks.
3.3.2 Proper amount of additives
Xylan from beechwood (0.85g, dry weight) was dissolved in 90 g DI water by heating at
69±1℃ for about 30 min with magnetic stirring. After cooling down to room temperature,
different amounts of sorbitol and glutaraldyhyde (GA) were added to the xylan solution at a
constant ratio of 4:1 (sorbitol: GA). A series of solutions with different total additive amounts of
0.125, 0.25, 0.5 and 0.75 gram were obtained. A xylan solution without additives was the control.
Each of these solutions was then magnetically stirred, transferred to 3 petri dishes after screening,
evaporated to form films and conditioned as described above.
3.4 Hemicellulose extraction and film fabrication
Before hemicellulose extraction, it was necessary to remove extractives and ash so that
they did not cause any interference with the further treatment. Lignin was removed as well for
the bleached hemicellulose samples.
3.4.1 Extractives removal
Due to health hazards associated with the use of benzene, ethanol and hot water were
used as the solvents to remove the extractives in wood, replacing the ethanol-benzene mixture
used previously in accordance with NREL/ TP-510-42619. The Soxhlet method was used here.
Oven dried hybrid poplar powder (10 g) was put in a pre-weighed extraction thimble and
inserted to the Soxhlet tube. HPLC grade water (190±5 ml) was added to the tared receiving
flask and adjusted to provide about 5 siphon cycles per hour and refluxed for 24 hours; the same

38

amount of 190 proof ethanol (190±5 ml) was used subsequently with about 10 siphon cycles per
hour and refluxed for an additional 24 hours. The ethanol mainly removed wood resin material,
including fatty and resin acids, sterols, waxes and fats, which can be dissolved in organic
solvents. Hot water removed the water-soluble compounds such as polyphenols and low
molecular weight carbohydrates, for example, tannins, gums, sugars, starches, and coloring
matter. Some of the low molecular weight hemicellulose could be removed at the same time.
Meanwhile, some ash was removed in the hot water extraction procedure, also.

3.4.2 Metal ion chelation
The raw materials normally contained metal ions (such as Mn2+ and Fe2+) that could
catalyze the disproportionation of hydrogen peroxide and interfere with the bleaching procedure.
For example, soluble ferrous iron can be rapidly oxidized by hydrogen peroxide to ferric (Fe3+),
forming a rapidly settling ferric hydroxide floc. Deactivation of the metal ions is thus necessary.
Ethylenediaminetetraacetic acid (EDTA) can inhibit the ability of metal ions to catalyze the
disproportionation of hydrogen peroxide. Once bound to EDTA, these metal centers tend not to
form precipitates or to interfere with the action of hydrogen peroxide. Therefore,
ethylenediaminetetraacetic acid was used to bind those ions to eliminate their activity. Each raw
material was treated with 0.2% EDTA (wood:EDTA ratio = 1:30, w/w) in a plastic bag placed in
a water bath at 70

for 1 hr. After that, the raw materials were filtered, and washed with DI

water.

39

3.4.3 Bleaching
A bleaching solution containing 20% sodium percarbonate (SPC), 0.05% magnesium
sulfate and 3% sodium silicate was used to remove lignin from the extractive-free hybrid poplar
powder, aspen CTMP and spruce CTMP. Sodium percarbonate is in fact a carbonate perhydrate
with chemical formula, 2 Na2CO3 • 3 H2O2. It is a colorless, crystalline, hygroscopic and watersoluble solid. It is a source of highly concentrated hydrogen peroxide in non-aqueous systems,
even if the material is not fully soluble. The addition of small amounts of water or sonication can
increase the rate of oxidation. Dissolved in water, it yields a mixture of hydrogen peroxide
(which eventually decomposes to water and oxygen) and sodium carbonate. It is very storagestable if it is dried. So it is much safer to store or transport compared with liquid hydrogen
peroxide. Furthermore, it contains no phosphorus or nitrogen; it is an eco-friendly chemical, and
has been used as an ingredient in a number of home and laundry cleaning products. Thus, sodium
percarbonate was selected as the bleaching agent in this study. EDTA treated raw materials with
bleaching solution (1:50, w/w) were placed in a plastic bag kept in a water bath at 85-90

for 1

hr. The raw materials were then filtered and washed with DI water.

3.4.4 Alkaline extraction
After the treatment procedures above, the leftover material was mainly hollocellulose,
which contains cellulose and hemicellulose. Potassium hydroxide with a concentration of 10%
(w/w) was used as the alkaline extraction solution to extract hemicellulose from the
hollocellulose at a sample to solution ratio of 1:50 (w/w). This extraction was conducted at room
temperature for 20 hours with magnetic bar stirring. Then the supernatant liquor was collected by
centrifugation (Eppendorf, 5804R, Germany) at 4500 rpm for 4 min for further treatment.

40

3.4.5 Hemicellulose recovery
The ethanol method was applied to precipitate the hemicellulose from its alkaline
solution at a ratio of 1:1 (190 proof ethanol, w/w). The precipitate was collected by
centrifugation and washed with 1:1 water-ethanol solution twice with repeated centrifugation.
190 proof ethanol (95%) was used for the final wash to remove the low molecular weight organic
matter. Then, the solid remainder was dried at room temperature for film making. The used
ethanol was recycled by rota-evaporation.

3.4.6 Fabrication of extracted hemicellulose films by solvent casting
One gram of hemicellulose was dissolved in 90 g DI water by heating at 69±1

for

about 15 mins. After the temperature dropped to room temperature, 0.2 g sorbitol (plasticizer)
and 0.05 g glutaraldehyde (cross-linking agent) were added into the solution. The solution was
magnetically stirred for 5 min, transferred evenly to 3 petri dishes (100 mm diameter × 15 mm
height) by passing through a 230 mesh screen and evaporated at room temperature to form films.
A digital micrometer was used to measure the thickness of the films. For each approach, six
films were made. Hemicellulose films without additives and with the same amount of sorbitol
were fabricated as well for comparison.
3.5 Characterization and property analysis
3.5.1 HPLC analysis
Sugar analysis of the hemicellulose was conducted using high performance liquid
chromatography (HPLC). HPLC is a technique to separate, identify and quantify the different
components in a mixture. The typical detectors used for HPLC include UV-VIS detector, photo

41

diode array detector, fluorescence detector, mass spectroscopic detector, refractive index detector
and light scattering detectors.
The composition analysis of the hemicellulose obtained from the six different approaches
in this study was carried out according to modified protocol NREL/TP-510-42618. A 0.1 g
hemicellulose sample was first hydrolyzed in 29 mL 4% sulfuric acid solution by placing in an
autoclave at 121˚C for 1 hr. The hydrolysate was then neutralized by calcium carbonate until no
bubble production. The sugar content after neutralization was determined by HPLC equipped
with refractive index (RI) detector and an Aminex HPX-87P 9 μm, 7.8 x 300 mm column (BiRad Laboratories, Inc., CA, USA) kept at 40˚C. An isocratic elution was carried out at a flow
rate of 0.6 mL min-1 for 17 min with a mobile phase of DI water. Acetyl groups in the
hemicellulose molecular chain were determined by HPLC with an Aminex HPX-87H 9 μm, 7.8
x 300 mm column (Bi-Rad Laboratories, Inc., CA, USA) kept at 40˚C. A mobile phase of 6M
sulfuric acid solution was used to elute the sample at the same flow rate as in the sugar content
measurement. The other conditions were the same. All the samples were filtered by a 0.45 μm
membrane (Baker) before injection into the HPLC. The solid residue in the acid solution was
collected by paper filtration and dried for measurement of lignin content as well as ash content
contained in the extracted sample after burning. All the tests were conducted in duplicate.

3.5.2 SEC measurement
Size exclusion chromatography was used to estimate the molecular weight of the
extracted hemicellulosic preparations using elution in an aqueous mobile phase solvent
containing 0.1 M sodium nitrate and 0.01 M sodium hydroxide. The molecules with different
molecular weights were separated based on their size during their elution through the column and

42

detected using a refractive index (RI) detector. The chromatogram equipment included two
Waters (Milford, MA) columns (Ultrahydrogel 250, 7.8 mm X 300 mm and Ultrahydrogel 500,
7.8 mm X 300 mm) to improve the separation of hemicellulose molecules.
Dextran standards with molecular weights of 6000, 40000, 70000, 100000 and 2000000
g/mol were used to calibrate the columns. The extracted hemicellulosic preparations were
dissolved in an aqueous solution similar to the mobile phase at a ratio of 1:100 (solid: liquid
mobile phase) through heating for 60 minutes at a temperature of 66±2 ℃ with sonication as
well. Sonication was used to reduce the aggregation of hemicellulose molecules during
dissolution (Stoklosa and Hodge, 2012).
The SEC column was kept at a temperature of 40 ℃ and each sample (injection volume:
20μl) was run for 48 min with a flow rate of 0.6 ml/min. All the samples were run in triplicate.
Knowing the elution time and the flow rate, the elution volume was calculated by
multiplying the flow rate by the elution time as expressed in the equation below:
Volume=Retention time ×flow rate
Number average molecular weight (Mn), weight average molecular weight (Mw) and
polydispersity index (PI) of hemicellulose samples based on the calibrated elution volume range
were calculated from the SEC analysis results.
The number average molecular weight is the statistical average molecular weight of all
the polymer chains in the sample, defined as:
∑
∑
where Mi is the molecular weight of a chain, Ni is the number of chains of that molecular weight.
The weight average molecular weight takes into account the molecular weight of a chain in
determining contributions to the molecular weight average. It is defined as:
43

∑
∑
In the SEC approach, molecular weight distribution information is processed by dividing the
output from the concentration detector into a number of time slices of equal width. Each of these
time slices stands for a type of monodisperse molecular weight based upon the calibration curve.
Next, the corresponding elution volume, molecular weight (Mi), area (Ai), cumulative area,
cumulative area percent, Ai divided by Mi, Ai times Mi, and the summary of the last two factors
are calculated. These calculated data are used to compute the molecular weight averages and
distributions (Malawer and Senak, 2003).

Mn

Mw

∑
∑

(1)

⁄

∑

(2)

∑

The polydispersity index is used as a measure of the broadness of a molecular weight distribution
of a polymer, defined as:

The larger the polydispersity index, the broader the molecular weight.
The same measurement was carried out for xylans from birchwood, beechwood and oat
spelts. The number average molecular weight and weight average molecular weight were
converted to DP later by conversion factor of 132 g/mol monomer subunit for a xylan polymer
for comparison with the results obtained by the reducing-end method.
3.5.3 DP measurement by reducing-end method
The reducing-end method provided the number average degree of polymerization of these

44

xylans, which was determined as the ratio of xylosyl monomer concentration divided by the
reducing-end concentration. The xylosyl monomer concentration was determined by the same
4% sulfuric acid hydrolysis method discussed in the composition determination of extracted
hemicellulosic samples. The reducing-end concentration was determined by a 2, 2’bicinchoninate (BCA) method (Zhang and Lynd, 2005). This method featured incubation of the
mixture of xylan solution and BCA working solution at 75 °C for 30 min. The BCA working
solution was made by mixing equal volumes of two solutions. The first solution contained 0.971
g of disodium 2, 2-bicinchoninate, 27.14 g of Na2CO3, and 12.1 g of NaHCO3 dissolved in 500
mL of distilled water. The second solution contained 0.624 g of CuSO4‚5H2O and 0.631 g of Lserine dissolved in 500 mL of water. Then absorbance was measured at 560 nm. Solutions of
xylose with concentrations of 0, 5, 10, 20, 40 and 60 µM were run as standards. Since it is well
known that the (4-O-Me)-GlcAp subunits are substituted onto the polymer backbone in xylans as
an α-1, 2 glycosidic bond (Teleman et al, 2002) and the Araf subunits are linked to xylans with α1, 2 linkage and/or α-1, 3 linkage (Ebringerov , 2005), there should only be one unique anomeric
carbon or ―reducing-end‖ per polymer chain in the xylans. Thus, the reducing-end concentration
could be obtained. In combination with the xylosyl monomer concentration, the number average
degree of polymerization of the xylans from different sources could be estimated.
3.5.4 Enzyme hydrolysis of hemicellulose
Hemicellulose (1 g) obtained from bleached hybrid poplar powder (HPPB) was put into a
vial, and 10 g DI water added. Then the pH of the liquid was adjusted to 4.5-5.0 by sulfuric acid.
A buffer solution (pH 5.5, 650 ml) was added to the liquid to maintain the pH during hydrolysis.
Then DI water was added to make the liquid reach 13 g total.

45

Enzyme HTec 2 (Cellic, Novozymes, Bagsvaerd, Denmark) 61µl was added to this final
liquid to reach a 1% concentration of enzyme (enzyme/hemicellulose, w/w). Then the vial was
put in an incubator and shaken for 1 hour at a temperature of 50℃.
After hydrolysis, the vial was put in boiling water for 5 minutes to kill the enzyme. The
solid was collected by centrifugation. Seventy percent ethanol (190 proof) solution was used to
wash the solid for 4 times with repeated centrifugation. Then the solid was dried at room
temperature for SEC measurement. Later, Mn and Mw obtained by SEC were converted to DPs
by conversion factor of 132 g/mol monomer subunit for a xylan polymer for comparison.
3.5.5 FTIR analysis
Fourier transform infrared spectroscopy (FTIR) is a technique to identify the typical
functional groups in a material and semi-quantification could be conducted based on the different
absorption of infrared caused by the motion of the functional groups. It was used to identify
lignin in both bleached and unbleached hemicellulose samples in this study. The samples from
both bleached and unbleached HPP hemicellulose with the same weight were scanned by a
Shimadzu IRPrestige-21 FTIR apparatus (Shimadzu, Japan) equipped with a Pike Technologies
horizontal attenuated total reflectance (HATR) accessory. Infrared absorbance spectra within a
wavenumber range of 2500-600 cm-1 were obtained with resolution number of 4 and scan
number of 60. Two lignin samples, 42-7 bmwel lignin and AlcellTM organosolv lignin (Repap
Inc., Valley Forge, PA, USA) were scanned as the lignin determination standards. Xylan from
oat spelts (Sigma-Aldrich, USA) and hybrid poplar powder were also scanned as control samples.
The experiment was conducted in triplicate.

46

3.5.6 UV-Vis analysis
Compounds containing double/triple bonds and nonbonding electrons can absorb ultraviolet
radiation or visible radiation. So a Lambda 25 UV-Vis spectrometer (Perkin Elmer Instruments,
Waltham, Massachusetts, USA) was used to record the UV-Vis spectra of HPP hemicellulose
based films in the range of 200-800 nm to identify and determine lignin in the hemicellulose
based films. Meanwhile, information about the transparency of these hemicellulose based films
was obtained.

3.5.7 XRD analysis
X-ray diffraction (XRD) was employed to explore the crystal region of the hemicellulose
based films if any formed during casting. It is a tool used for determining the atomic and
molecular structures of crystal lattices in a sample without destruction. A crystal can diffract a
beam of incident X-rays into many specific directions. By measuring the angles and intensities of
these diffracted beams, the density of electrons within the crystal, as well as the mean positions
of the atoms and their chemical bonds, their disorder etc., can be obtained.
In this study, X-ray diffraction was used to characterize the structure of the extracted
hemicellulose powder as well as hemicellulose films with/without additives. The XRD
instrument was a Bruker AXS D8 Advance X-ray diffractometer (Bruker Co., Billerica, MA,
USA) equipped with a Cu Kα radiation source (wavelength λ = 0.154 nm) setting at 40 kV and
40 mA. The sample was scanned over a 2 theta range of 10°to 45°at a step time of 0.5 s and a
step size of 0.02°. Sorbitol was also scanned as a control.
Xylan based films using xylans purchased from Sigma-Aldrich (S: G=4:1, 25% additives)
were scanned in the same manner.

47

3.5.8 DSC analysis
Differential scanning calorimeters (DSC) measure the relationship between temperatures
and heat flows, which can indicate the thermal transitions in a material, such as glass transition,
―cold‖ crystallization, phase changes, and melting. By knowing this information, the structure
and thermal stability of the material can be obtained.
In this study, a DSC Q 100 (TA Instruments, New Castle, DE, USA) was employed to
record the spectra of HPP bleached and unbleached hemicelluloses and their films. These films
were made from a series of recipes (hemicellulose with no additives; hemicellulose:sorbitol
/1:0.2;

hemicellulose:sorbitol:glutaraldehyde

/

1:0.2:0.05).

Before

measurement,

these

hemicellulose samples were dried at a temperature of 50°C for 4 hrs. These samples were all
measured over a temperature range of 42

to 280

at a heating scan rate of 10

min-1.

3.5.9 Tensile tests
For the mechanical properties analysis, an Instron electromechanical load frame (series
5565) was used to test the tensile properties. Based on the tested data, tensile strength (TS),
elongation at break (El), modulus of elasticity (MOE), specific strength and flexural stiffness (S)
were obtained.
Tensile tests are fundamental materials tests in which a sample is subjected to a
controlled force until failure. A complete tensile profile can show how the tested sample reacted
to the forces being applied. Information about properties such as tensile strength, elongation at
break, tensile energy absorption (TEA), Young's modulus, tensile stiffness and yield strength can
be obtained. Results are typically displayed in a schematic as shown below from Instron (Figure
11).

48

Figure 11 Tensile stress-strain diagram
For most tensile testing of materials, in the initial portion of the test, the curve of the
profile is linear. In this linear region, the line obeys the relationship defined as "Hooke's Law"
where the ratio of stress to strain is a constant, or

. E is the slope of the line in this region

where stress (σ) is proportional to strain (ε) and is called the "modulus of elasticity" or "Young's
modulus". The modulus of elasticity is a measure of the stiffness of the material, but it only
applies in the linear region of the curve. If a specimen is loaded within this linear region, the
material will return to its exact same conditions if the load is removed.
The specific strength of a material is calculated by the material's strength divided by its
density. This means not only the strength of a material is considered, but also at the same time,
the material’s weight is considered. It is an important way to remove the influence of the
material’s density. The material density is calculated by the mass divided by the volume. The
49

strength can be obtained from the tensile properties test. A high specific strength can mean a
material has both lighter weight and higher strength.
As a packaging material, the bending stiffness of a film is an especially important quality
parameter. It affects the appearance, machinability and functionality of the product (Koran and
Kamdem, 1989; Wyser et al., 2001). The bending stiffness, also referred to flexural stiffness (S),
is defined as the resistance of a sheet material to bending forces. Many different instruments
have been designed to measure the flexural stiffness using different measurement schemes, such
as the Taber tester, Lhomary tester, Gurley tester, Instron tester and Clark tester (Koran and
Kamdem, 1989). The Instron tension method is a relatively simple approach to measure stiffness
in theory:

where E is the elastic modulus, b is the width of the material and I is the moment of inertia of the
cross-section. For a uniform film, the equation above can be modified by applying the moment
of inertia of a rectangular cross-section:

where t is the thickness of the material. Hence:

The flexural stiffness is finally proportional to the tensile modulus multiplied by the thickness to
the third power. Thus, the flexural stiffness depends on both the intrinsic property of the material
(E) and its thickness (t). For very thin films, the flexural stiffness will be very low. Even though
the Instron tension method for measurement of the flexural stiffness is rather simple in theory, it

50

has too low sensitivity to permit accurate determination of this property for a film material
(Wyser et al., 2001).
In this study, the samples from the six approaches were conditioned at about 23

and

RH 50% until equilibrium before tensile properties testing. The moisture content was measured
by the weight method, drying at 105

for 5 hours. Then samples were cut into rectangular strips

of size 2‖ 0.25‖. Thickness was measured before testing using a TMI digital micrometer at
three points for each strip, and an average was calculated. Twenty four replicate specimens
equally from six films were tested using an Instron universal testing machine (Instron 5565) with
a load cell of 100 N. The samples were tested with an initial grip separation of 1 inch and
stretched at a speed of 0.5 inch/min. A computer was used to record the force-extension curve.
Tensile strength in psi and maximum load were output by the computer. Elongation in
percentage was defined as the percentage change in the length of the specimen compared to the
original testing length. The modulus of elasticity was defined as the ratio of stress to strain at the
proportional limit.
The tensile tests for hemicellulose films with smaller thickness values and different
additives from the HPP approach were conducted under the same conditions, except that 8
replicates were applied in these tests.
The same test was applied to the purchased xylan based films (S: G=4:1, 25% additives)
and also the films (S: G=4:1, 25% additives) made of hemicellulose samples from HPPB
with/without enzyme hydrolysis.
3.5.10 SEM analysis
Scanning electron microscopy (SEM) was used to display the surface morphology of the
hemicellulose based films. SEM is a type of electron microscope that produces images by

51

scanning the sample with a focused beam of electrons. The electrons interact with atoms in the
sample surface, and then the atoms emit secondary electrons which were excited by the electron
beam. When they are captured and detected, they contain information about the sample's surface
topography and composition which is displayed in image form. SEM can achieve resolution
better than 1 nanometer.
In this study, bleached and unbleached hemicelluloses extracted from hybrid poplar
powder based films were cut into small pieces and mounted on aluminum stubs using high
vacuum carbon tabs (SPI Supplies, West Chester, PA). These pieces were then coated with
osmium (≈10 nm thickness) in an NEOC-AT osmium coater (Meiwafosis Co., Ltd., Osaka,
Japan). A JEOL JS-6610LV scanning electron microscope (JEOL Ltd., Tokyo, Japan) was used
to examine these film samples.
3.5.11 Water vapor barrier property test
Mocon equipment was used to test water vapor barrier properties of the hemicellulose
based films. It allowed samples to be exposed under controlled relative humidity and temperature.
When transmission reached the steady state, the amount of the substance permeating the sample
in a unit time was measured. In this study, a water vapor permeation analyzer (Permatran-W
Model 3/33, MG module, Mocon) equipped with infrared sensor with parts-per-million
sensitivity was used to test the water vapor transmission rate (WVTR) at 23℃, RH 100% for 24
hours with a nitrogen carrier. It was conducted according to the standard ASTM F-1249, and
reported in units of g/ (day∙m2). At least six replicate samples were tested at each condition. The
permeability was calculated from the WVTR and the thickness measured separately, and
reported in units of g•mil/ [d•m2•mmHg].

52

3.5.12 Oxygen barrier property test
Similar Mocon equipment was used to test oxygen barrier properties of the hemicellulose
based films. An oxygen permeation analyzer (Ox-Tran Model 2/21, MH Module, Mocon)
equipped with infrared sensor with parts-per-million sensitivity was used to test the oxygen
transmission rate (OTR) at 23℃, RH 0% for one week with a nitrogen carrier. The test was
conducted according to the standard ASTM F-3985, and reported in units of cm3/ (day∙m2). At
least six replicate samples were tested at each condition. The permeability was calculated from
the OTR and the thickness measured separately, and reported in units of cm3•mil/ [d•m2•mmHg].

3.6 Statistical analysis
In order to analyze the difference between the six levels of treatment (CTMP aspen
bleached/unbleached,

CTMP spruce

bleached/unbleached

and

hybrid

poplar

powder

bleached/unbleached), and their effects on the tensile properties of hemicellulose films (tensile
strength (TS), elongation (El) and modulus of elasticity (MOE)), multivariate analysis of
variance (MANOVA) was conducted to analyze the collected data by different models; while
one-way analysis of variance (ANOVA) was applied also to analyze the effect of the six levels of
treatment on hemicellulose molecular weight, barrier properties and other experiments for
determining significant differences. Post-hoc HSD (honestly significant difference) pairwise
comparisons were conducted using the Tukey test with 95% level of confidence (α=0.05) in both
analyses (MANOVA and ANOVA). All analyses were conducted using SPSS (version 22, IBM
Corporation, NY, US).
In the statistical analysis of MANOVA, T1 represents bleached CTMP aspen; T2 represents
unbleached CTMP aspen; T3 represents bleached CTMP spruce; T4 represents unbleached

53

CTMP spruce; T5 represents bleached hybrid poplar powder; and T6 represents unbleached
hybrid poplar powder.

54

4 PROPERTIES OF XYLAN BASED FILMS AND EXTRACTED HEMICELLULOSE
BASED FILMS
4.1 Xylan based films
4.1.1 Function of additives
As is seen in Figure 25 (appendix), among these different ratios of sorbitol to GA (3:1,
2:1, 1:1, 1:0, 0:0, 0:1, 1:2, 1:3), only the film at the ratio of sorbitol to GA 1:0 could be peeled
from the petri dish. It formed a big piece of self-standing film, but was very brittle. No
continuous self-standing films were observed at the other ratios. At the ratios with higher sorbitol
such as 2:1 and 3:1 (sorbitol: GA), the films were soft but at the same time, sticky. They could
not be peeled from the petri dish as continuous big piece of films. The films at the ratios with
higher GA (0:1, 1:2, 1:3) showed similar behaviors as the films at the ratios with higher sorbitol.
They could not form continuous and self-standing films either. Comparable to the films with
higher sorbitol, they were much stickier at the ratios of sorbitol to GA 1:2 and 1:3 (sorbitol: GA).
The solutions with ratios of 1:1 and 0:0 (sorbitol: GA) formed self-standing films but noncontinuous, in a form of very brittle small pieces. The film with a ratio of 0:0 formed much
smaller brittle pieces compared to the film with a ratio of 1:1. This indicates that both of the
additives promote xylan to form a continuous film. But a proper ratio of sorbitol to
glutaraldehyde exists, too much or too little of either additive can lead to poor film formation
properties, such as sticky or brittle.

4.1.2 Proper amount of additives
As is seen in Figure 26 (appendix), among these different additive amounts, the solutions
with 0.125 g and 0.25 g additives showed better film forming ability than the other conditions.

55

They formed big piece of self-standing films, and could be peeled from the petri dish. But the
film with 0.125 g additive was more brittle than the one with 0.25 g. No continuous self-standing
film was observed under the other conditions. With higher additive addition, the xylan films
tended to be soft and sticky. They could not be peeled from the petri dish as continuous big piece
of films. The film with 0.25 g (Sorbitol: GA=4:1) additive displayed the best film properties,
continuous and flexible, under the conditions tested in this study.

4.2 Extracted hemicellulose based films
4.2.1 UV-Vis spectra of hemicelluloses based films
The UV-Vis spectra of bleached and unbleached HPP hemicelluloses based films were
shown in Figure 12. The absorption spectrum is to display from 260nm to 400nm (Figure 12 a),
the transmittance spectrum is shown from 200nm to 800nm (Figure 12 b). As is seen from the
absorption spectrum (Figure 12 a), unbleached HPP hemicellulose film shows higher absorbance
than bleached HPP hemicellulose film in the whole range of 260nm-400nm. There is an obvious
absorption peak at 275 nm on the curve of unbleached HPP hemicellulose film, which is the
characteristic absorption maximum corresponding to the π →π* electron transition in the
aromatic ring of lignin. This has a good agreement with the research on bark lignin (Seca et al.,
1998). This type of electron transition usually appears in lignin at 280 nm. The hypsochromic
shift is due to the relatively higher content of syvingyl unites in the lignin (Morohoshi, 1991;
Seca et al., 1998). The absorption band above 300 nm indicates the presence of structures
containing unsaturated moieties conjugated with the aromatic moieties in lignin (Oliveira et al.,
2006; Singh et al., 2012).

56

The transparencies of both of the two samples increase from the UV radiation to
visible radiation (Figure 12 b). This result has a good agreement with the study of hemicellulose
film reported by Guan et al. (Guan et al., 2014). The HPP bleached hemicellulose based film
increased more sharply compared to the unbleached sample. Moreover, the transparency of HPP
bleached hemicellulose based film was higher than that of unbleached film in the range of UV to
visible region (300-800 nm). It was reported that the absorbance in the visible region was due to
the chromophoric groups contained in lignin (Singh et al., 2012). These results indicate that
lignin plays an important role on the absorption of the UV and visible light in the hemicellulose
based films. It reveals that unbleached HPP hemicellulose film contains higher lignin content
than bleached HPP hemicellulose film. This result corresponds to the result of composition
analysis; the lignin contents in the bleached samples were slightly smaller than that in the
unbleached samples. Meanwhile, the unsaturated bonds and n electrons from atom O can be
excited by UV and visible radiation (Robinson et al., 2005). Absorption then might come from
the C=O bond contained in the additive compound, glutaraldehyde. However, when same
amount of glutaraldehyde were added into both of the bleached and unbleached films, no
difference could be considered between the two samples from this point of view. So it is
confirmed that the changing of the absorbance and transmittance spectra results from the
differences of lignin content in the two types of samples. The transmittance of the two types of
films in the UV and visible ranges favors the packaging application, where the packed item is
designed to be seen by the customers, especially the bleached sample, which was over 80%. The
visual colors of the unbleached hemicellulose films in this study were all yellowish than the
bleached films, which was in agreement with the UV-Vis analysis results.

57

7

(a)
6

Absorbance

5
4

Unbleached HPP
hemicellulose film

3
2
1

Bleached HPP
hemicellulose film

0
260

280

300

320

340

360

380

400

Wavelength (nm)
100

(b)

Bleached HPP
hemicellulose film

Transmittance (%)

80

60

40

Unbleached HPP
hemicellulose film
20

0
200

300

400

500

600

700

800

wavelength (nm)

Figure 12 UV-Vis absorption spectrum (a) and transmittance spectrum (b) of bleached and
unbleached HPP hemicelluloses based films

58

4.2.2 Thermal properties of hemicellulose based films
The DSC spectra of bleached & unbleached extracted hemicelluloses (BH & UH) and
their films without additives (BHF & UHF), with sorbitol (BHSF & UHSF), with sorbitol and
glutaraldehyde (BHSGF & UHSGF) are displayed in Figure 13. These spectra present two clear
endothermic peaks for all of the bleached and unbleached HPP hemicellulose powder and their
films. The first endothermic peak behavior is similar to the study of extracted hemicellulose from
olive, chilli and pepper (Egüés et al., 2013).

This endothermic peak for unbleached HPP

hemicellulose is observed around 114 ℃. After it formed film, it became more thermal stable
than hemicellulose powder. The peak point increased from 114 ℃ to 140 ℃ for the unbleached
hemicellulose film with no additives. However, the addition of the additives made the films less
thermal stable. The temperature for this endothermic peak point decreased to around 126 ℃ for
unbleached sorbitol-hemicellulose film and sorbitol-glutaraldehyde-hemicellulose film. The Tg
values are observed to shift to a lower temperature (80-90 ℃ to 70-75 ℃) with addition of the
two additives. The plasticizer increased the chain mobility which reduced Tg, increasing the film
flexibility (Georget et al., 1999; Matveev et al., 2000). The endothermic peak that appeared at
around 240-250 ℃ for unbleached hemicellulose films without additives is suggested to be
attributable to the melting of the crystallites. When the unbleached hemicellulose formed film,
this peak changed to be sharper. This obvious change indicates a difference in the structure of the
material. Combined with the results from XRD analysis, it is probably due to the crystalline
melting. It is worthy to notice that two melting peaks are present in this endothermic event. Sato
et al. suggested two types of crystalline formation from solvent induced crystallization (Sato et
al., 2013). Since other research reported hemicellulose weight loss between 220 ℃ and 315 ℃
(Gao et al., 2013; Xiang et al., 2014), it may also be due to hemicellulose decomposition. It was
59

also reported that addition of plasticizer could increase the spacing between the crystal lattice
planes (Gröndahl et al., 2004; Mikkonen et al. in 2009). Thus the structure of the crystallites
changed. It is shown on the changing of the endothermic event at around 240 ℃ - 250 ℃ for
unbleached sorbitol plasticized hemicellulose films compared with the unbleached hemicellulose
film without sorbitol. It is worth noticing that the two endothermic peaks remain. Similar results
are found in the bleached hemicellulose and its films, except that only one endothermic peak
presents at around 240 ℃ - 250 ℃ after the addition of plasticizer. The exact reason for this
phenomenon is not known. Bleaching treatment is a complex process.

60

a

BHSGF

Endothermal heat flow (W/g)

BHSF

BHF

BH

0

50

100

150

200

250

300

Temperature (℃)

Figure 13 DSC thermograms of HPP hemicelluloses and their films with and without additives,
(a) bleached; (b) unbleached.

61

Figure 13 (cont’d)
b

Endothermal heat flow ((W/g)

UHSGF

UHSF

UHF

UH

0

50

100

150

200

250

300

Temperature (℃)

4.2.3 Mechanical properties of hemicellulose based films
Data for tensile strength (TS), modulus of elasticity (MOE), percent (%) elongation at
break (El), and other physical and mechanical properties for different hemicellulose based films
are listed in Table 2. And also, a typical tensile load-extension diagram obtained in this study is
displayed in Figure 14.
The highest TS is from unbleached HPP hemicellulose based films, which is 76.5 MPa. This was
about 60% higher than the TS of films made from arobinoxylan obtained from corn hull

62

plasticized with sorbitol (Zhang and Whistler, 2004), and even higher than the strength of
sorbitol plasticized birch wood xylan films enhanced with nanofibrillated cellulose which was 66
MPa when plasticized with 20% sorbitol (Hansen et al., 2012). It was reported that the tensile
strength of xylan films varied from less than 1 MPa with nearly 90% elongation to 50 MPa with
2.5% elongation (Mikkonen et al., 2009). These properties are highly related to hemicellulose
source, concentration & type of plasticizer, molecular weight, and substitution (Gröndahl et al.,
2004; Mikkonen et al. in 2009; Heikkinen et al., 2013; Heikkinen et al., 2014). For hardwood
xylan film with 20% plasticizer, such as sorbitol and glycerol, elongation is below 2%; for oat
spelt xylan film, it is above 5%. But oat spelt film tensile strength is lower than for hardwood
xylan film (Gröndahl et al., 2004; Mikkonen et al. in 2009). In this study, the largest elongation
comes from the films made from bleached HPP (7.4%), larger than the other hardwood xylan
films with a similar amount of sorbitol (less than 2%) (Gröndahl et al., 2004). And unbleached
HPP hemicellulose based films also provide the biggest MOE value (3.5 GPa), which is superior
to herbaceous plant xylan film (0.5 GPa-1.0 GPa), similar to hardwood xylan film (close to 4.0
GPa) (Gröndahl et al., 2004; Mikkonen et al. in 2009; Heikkinen et al., 2013; Heikkinen et al.,
2014). Compared with cellophane, a successful commercialized cellulose film, the TS of
hemicellulose films obtained in this study is much lower (76.5 MPa in this study versus 101.6
MPa), but comparable elongation at break (7.4 % in this study versus 7.6%), and higher MOE
(3.5 GPa in this study versus 1.3 GPa) (Bhat and Makwana, 1988). However, the tensile strength
values of films in this study are much larger than the commonly used packaging petro-plastic,
high density polyethylene (HDPE), which is 29.9 MPa. The highest MOE of these films is 2.3
times that of HDPE. But the elongation at break is one order of magnitude lower. For convenient
comparison, the three main tensile properties of the current most-used plastics in packaging are

63

listed in Table 3. As is seen from this table, the TS of the commonly used plastics are typically
below 50 MPa, much smaller than the highest TS of the films in this study.
The specific strength showed the same tendency as tensile strength, which removed the
influence of the density on the tensile strength of the films. The flexural stiffness followed the
direction of the sample thickness. The thickness of the film sample had a very strong effect on
the flexural stiffness, since the MOEs of the six samples in this study were comparatively close.
From this point of view, it is suggested that the bending stiffness of the hemicellulose based
films in this study is comparable or better than that of common commercial packaging plastics.

6
5

Load (lbf)

4
3
2
1
0
0

0.005

0.01

0.015

0.02
0.025
Extension (in)

0.03

Figure 14 Tensile load-extension diagram

64

0.035

0.04

Table 2 Mechanical properties of hemicellulose films from different treatments
CTMP Aspen
Bleached

CTMP Aspen
Unbleached

CTMP Spruce
Bleached

CTMP
Spruce
Unbleached

HPP
Bleached

HPP
Unbleached

Tensile strength, MPa

47.0 (3.8)

51.8 (4.6)

59.6 (5.5)

71.0 (7.3)

66.8 (3.7)

76.5 (7.0)

29.9

Elongation, %

5.6 (1.5)

4.4 (0.6)

7.0 (1.3)

6.2 (1.4)

7.4 (1.1)

5.4 (0.9)

56.5

MOE, GPa

2.7 (0.3)

2.7 (0.2)

2.3 (0.3)

2.7 (0.3)

2.9 (0.2)

3.5 (0.4)

1.5

MC, %

14.5 (0.3)

14.4 (0.5)

13.0 (2.0)

14.5 (0.3)

15.9 (0.7)

17.0 (0.3)

Thickness, m

63.5 (7.6)

66.0 (10.2)

119.4 (20.3)

116.8 (12.7)

91.4 (10.2)

68.6 (5.1)

Density, g/cm3

1.4 (0.1)

1.3 (0.0)

1.4 (0.1)

1.4 (0.1)

1.2 (0.1)

1.3 (0.0)

Specific
strength,
N·m/kg, 103

33.6

39.8

42.6

50.7

55.7

58.8

Flexural
N•m, 10-5

5.8

6.5

32.6

35.9

18.5

9.4

Properties

stiffness,

Notes: The numbers in parentheses are STDEV.
All measurements were performed in 20 replicates.
HDPE: Tensile Tests of Various Plastic Materials (1). Available from:
http://www.shimadzu.com/an/industry/petrochemicalchemical/i215.html

65

HDPE

Table 3 Tensile properties of several commercially used plastics in packaging
Polymer Type

Density
(g/cm3)

Tensile strength
(MPa)

Elongation at
break (%)

High-density
0.95-0.97
20-30
10-1000
polyethylene (HDPE)
Low-density
0.92-0.93
8-30
100-650
polyethylene (LDPE)
Polypropylene (PP)
0.90-0.91
30-40
100-600
Polystyrene (PS)
1.0-1.1
35-50
1-2
Acrylonitrile-butadiene1.0-1.1
15-55
30-100
styrene (ABS)
Polyvinyl chloride
1.3-1.6
40-50
2-80
(PVC)
Notes: Properties and applications of commercially important plastics.
Available from: http://www.britannica.com/EBchecked/topic/463684/plastic.

Modulus of
Elasticity
(MOE) (GPa)
1-1.5
0.25-0.35
1.2-1.7
2.6-3.4
0.9-3.0
2.1-3.4

Statistical analysis was then conducted on the tensile properties of hemicellulose based
films. Three dependent varables (TS, El and MOE) were contained in this study. Thus,
multivariate analysis of variance (MANOVA) is suitable to use for assessing the difference of
the

six

levels

of

treatment

(CTMP

aspen

bleached/unbleached,

CTMP

spruce

bleached/unbleached and hybrid poplar powder bleached/unbleached) and their impacts on these
three tensile properties.
1) One-way MANOVA
Before processing MANOVA analysis, normality of the collected data was checked by Q-Q
Plot and correlations among the tensile properties were examined pairwise for feasibility of use
of MANOVA.
A normal Q-Q Plot of tensile strength of hemicellulose films from bleached aspen CTMP
pulp fiber was taken as an example (Figure 15). A nice match of observed value and expected
normal value is displayed in (a). The distribution of observed data does not exhibit a systematical
pattern (b). Similar results were observed in the other normal Q-Q Plot graphics (Figure 27,
66

appendix). The normality of the collected data is good enough to process multivariate ANOVA.
(a)

(b)

Figure 15 Normal Q-Q Plot of tensile strength of hemicellulose films from bleached aspen
CTMP pulp fiber
The correlation analysis (Table 13, appendix) indicated that all of the three outcomes
(tensile strength, elongation and MOE) were correlated pairwise with each other (p<0.05). From
Pearson analysis, the correlations between TS and El; TS and MOE are positive (0.187; 0.581),
which means tensile strength behaves similarly to elongation and MOE. But the similarity
between TS and EL (0.187) is much weaker than it between TS and MOE (0.581). Usually, only
when the correlation coefficient is larger than 0.6, a strong positive correlation can be concluded.
Yet the correlation between elongation and MOE is negative (-0.200), which means elongation
behaves in an opposite way comparing with MOE, but not strong.
From the above analysis, the collected data are distributed randomly and the three tensile
properties are significantly correlated pairwise with each other. MANOVA is approved to be
suitable for use in this study.
1) One-way MANOVA

67

One-way MANOVA is used to analyze the effect of the six levels of treatment on the
means of the three tensile properties. Pillai’s Trace test was chosen to determine statistical
difference across treatment levels because of its robust properties (Hair et al., 2009). The results
indicate that the six levels of treatment have a significant effect on the tensile properties of
hemicellulose films (p=0.000) (Table 14, appendix). To determine which of these three tensile
properties contributed to the significant difference, a univariate test was further conducted to
assess this effect. The outcome shows that the six levels of treatment significantly impact the
three tensile properties individually, TS, El and MOE (All the three P value equal 0.000) (table
15, appendix).
To assess how the six levels of treatment impact the means of the three tensile properties
individually, multiple pairwise comparisons among the six means for each property were
conducted post-hoc using the Tukey HSD test (Table 16, appendix). The results revealed that
nearly all of the tensile strengths were significantly different (p<0.05), except for that between
T12 (unbleached CTMP spruce) and T13 (bleached hybrid poplar powder) (p=0.133). For the
multiple pairwise comparisons of El and MOE, approximate half of the comparison groups
showed significant differences (p<0.05). Among all the six levels of treatment, the groups of TS
with the largest significant difference are T1 (bleached CTMP aspen) and T6 (unbleached hybrid
poplar powder). The largest difference value is -29.5465 with 95% confidence interval located
between lower bound -34.2487 and upper bound -24.8443. For elongation, the largest significant
difference presents between T2 (unbleached CTMP aspen) and T5 (bleached hybrid poplar
powder). The largest difference value is -3.0217 with the 95% confidence interval between lower
bound -3.9428 and upper bound -2.1006. In comparisons of MOE, the largest significant
difference exists between T3 (bleached CTMP spruce) and T6 (unbleached hybrid poplar powder)

68

with a difference value -1.1700, and 95% confidence interval in the range of -1.4364 to -0.9035.
More information about other pairwise comparisons can be obtained from Table 16.
2) Two-way MANOVA
The analyses above state how the six levels of treatment affect the tensile properties.
However, we are interested in the function of each of these factors, hardwood & softwood,
bleached & unbleached and chemithermomechanical pulp & mechanically ground powder. For
assessing the effect of each of the three factors on the tensile properties, further statistical
analysis needs to be conducted. Three-way MANOVA would be the ideal analysis method to
answer this question. However due to the scope of my research, collecting data from
bleached/unbleached softwood powder was not feasible. Therefore, two-way MANOVA was
applied to obtain the maximum information regarding the factors’ impact.
Within hardwood
Within hardwood, the effect of pretreatments (chemithermomechanical pulp &
mechanically ground powder; bleached & unbleached) on the tensile properties were assessed by
two-way MANOVA. The results indicate that both of the pretreatments (chemithermomechanical
pulp & mechanically ground powder; bleached & unbleached) had statistically significant
influence on the tensile properties (p<0.05) (Table 17, appendix), and on each of the properties
(TS, El and MOE) (p<0.05) (Table 18, appendix). They had significant interaction on the impact
of the three properties (p<0.05) (Table 17, appendix). The pretreatments did not show significant
interaction on elongation (p=0.224) but did on TS and MOE (p=0.013; p=0.000).
Within CTMP pulp
Within CTMP pulp, the impact of factors, hardwood & softwood and bleached &
unbleached, on the tensile properties were assessed by two-way MANOVA. Similar results were

69

found in two-way ANOVA within hardwood test. Hardwood & softwood and bleaching & nonbleaching worked in the same manner (Table 19 and Table 20, appendix).
These statistical analyses results indicate that the combination of hemicellulose sources
(hardwood & softwood; pulp & powder) and pretreatments (chemithermomechanical treatment
& mechanically ground; bleached & unbleached) highly affected the hemicellulose film tensile
properties. As is seen in the statistical analysis results, the three factors (hardwood & softwood;
CTMP treatment & mechanically ground; bleached & unbleached) all have statistically
significant impact on the three tensile properties (TS, El and MOE). Within hardwood, each of
the other two factors significantly impacts the properties individually. All of the TS from
unbleached samples were significantly bigger than that from the corresponded bleached ones; the
TS from mechanically ground samples were significantly bigger than that from CTMP ones
(p<0.05). Within CTMP pulp, the same impact was found. All of the TS from unbleached
samples were significantly larger than that from the corresponded bleached ones; the TS from
softwood samples were significantly larger than that from hardwood ones (p<0.05). Similar
impacts were observed for El and MOE in the two further MANOVA tests, only in the opposite
direction for El.
From Table 2, it can be seen that the hemicellulose based films from CTMP spruce and
were thicker than the other samples. To remove the influence of thickness, thinner hemicellulose
films obtained from three of the six approaches were made to compare with the thicker samples.
The results are listed in Table 4. In general, the mechanical properties of the thinner film
samples corresponded well to the thicker samples. No significant differences were observed
between thinner and thicker samples. Therefore, thickness did not appear to impact the observed
differences in mechanical properties between formulations.

70

Table 4 Comparison of mechanical properties of hemicellulose films with different thickness

CTMP Spruce Bleached

CTMP Spruce Unbleached

HPP Bleached

Thin

Thin

Thin

Thick

Thick

Thick

Tensile strength,
62.7 (1.3)a
59.6 (5.5)a
66.8 (3.6)b
71.0 (7.3)b
68.4 (3.1)c
66.8 (3.7)c
MPa
Elongation, %
6.5 (0.5)a
7.0 (1.3)a
5.6 (0.2)b
6.2 (1.4)b
6.6 (0.9)c
7.4 (1.1)c
MOE, GPa
2.5 (0.1)a
2.3 (0.3)a
2.6 (0.3)b
2.7 (0.3)b
2.9 (0.1)c
2.9 (0.2)c
MC, %
12.8 (0.8)a
13.0 (2.0)a
15.8 (3.6)b
14.5 (0.3)b
16.4 (0.3)c
15.9 (0.7)c
88.9 (2.5)
119.4 (20.3)
91.4 (7.6)
116.8 (12.7)
76.2 (5.1 )
91.4 (10.2)
Thickness, m
Note: The numbers in parentheses are STDEV.
All measurements for thin samples were performed in 8 replicates; for thick samples, there were 20 replicates.
Values between thin and thick samples in the same row in the same formulation by the same letter are not significantly
different (p 0.05).

71

4.2.4 Morphology of hemicellulose based films surface
a

Figure 16 shows the smooth evaporation surfaces
of both HPP bleached and unleached
b

hemicellulose based films, which revealed homogeneous structures on the surfaces. The small
white particles observed on the surface were probably attributable to the dust in the air or
impurities in the films, such as salt crystals. No nanocellulose-like substance was found in either
of the sample surfaces. This indicates that the mechanical property changes observed in the
tensile tests were attributable to the addition of sorbitol and glutaraldehyde, not to the
reinforcement by accidentally included nano particles, such as the nano cellulose fiber from the
hemicellulose extraction process (e.g. hemicellulose alkaline solution centrifugation).

Figure 16 SEM morphology of HPP hemicellulose based films, a: bleached; b: unbleached

4.2.5 Water vapor barrier properties of hemicellulose based films
The WVTR values of hemicellulose based films from the six different approaches are
listed in Table 5. WVP values were calculated based on the thicknesses measured separately. The
WVTR values of these films varied from 4600 g/ d•m2 to 5600 g/ d•m2, which were higher by
several orders of magnitude than those from current commercial packaging films (Table 6).
Among these plastics, HDPE is known as a common moisture resistant petro-polymer used in

72

food packaging. The WVTR values in this study are larger by 2-3 orders of magnitude than that
of HDPE. Furthermore, the WVP values obtained in this study are still two orders of magnitude
larger than the WVP value (7.9×10-11 g Pa-1m-1s-1) of CellophaneTM (Phan The et al., 2009).
Hence, the water vapor barrier properties of the films in this study need to be improved to meet
the requirements of commercial use in many food packaging areas.
As shown in Table 5, the WVTRs of unbleached hemicellulose based films were slightly
lower than those of bleached hemicellulose based films. This is probably attributable to the
higher crystallinity of unbleached hemicellulose based films, as shown in the XRD analysis for
hemicellulose based films from hybrid poplar powder. In addition, this trend corresponded to the
changes in tensile strength. In general, the higher was the tensile strength for the film, the lower
the WVTR value. This behavior is possibly attributable to the increased intermolecular bonding
in the films with higher tensile strength. The intermolecular bonding reduced the molecular
mobility in the hemicellulose matrix, and thus decreased the WVTR value. The WVTR values
obtained in this study were 2-3 times higher than the values (1800 g/ d•m2,10% sorbitol; 2050 g/
d•m2, 40% sorbitol) from arabinoxylan based films from oat spelt (Mikkonen et al., 2009),
which were reported to be comparable to the values from corn arabinoxylan films (Péroval et al.,
2002; Zhang and Whistler, 2004), methyl cellulose films (Greener-Donhowe and Fennema, 1993;
Phan The et al., 2002) and nanofibrillated cellulose reinforced birch wood xylan films (Hansen et
al., 2012), but they were the same level of magnitude. The reason why the WVTR values of
films in this study were higher than those from arabinoxylan based films from oat spelt may be
attributable to the different humidity gradients (0/100% in this study versus 0/86%),
measurement methods (Mocon method versus cup method) and also the different hemicellulose
sources. It was reported that the humidity gradient significantly affected the WVTR values of

73

hydrophilic films (Hagenmaier and Shaw, 1990; McHugh et al., 1993; Mikkonen et al., 2009).
The WVTR values for hydrophilic films can increase greatly at very high RH.

74

Table 5 Water vapor permeability of hemicellulose based films from six different approaches
Properties

CTMP
Aspen
Bleached

CTMP
Aspen
Unbleached

CTMP
Spruce
Bleached

CTMP
Spruce
Unbleached

HPP
Bleached

HPP
Unbleached

Thickness, m

63.5 (5.1)

58.4 (10.2)

137.2 (7.6)

124.5 (2.5 )

88.9 (5.1)

66.0 (10.2 )

5654 (180)a

5530 (246)a

4805 (363)b

4519 (356)b

4957 (281)b

4673 (151)b

122.3 (9.5)b

115.6 (15.2)ab

233.0 (12.6)e

200.0 (18.9)d

153.6 (4.8)c

107.5 (16.0)a

WVTR, g/(d•m2)
Permeability, 103

g•m/(d•m2•KPa)
Notes: The numbers in parentheses are standard deviation.
All measurements were performed in 6 replicates.
Values in the same row with different superscripts are significantly different (p 0.05).

75

Table 6 Water vapor transmission rates of common used plastic films
WVTR g/(d•m2)

Polymer type
Ethylene Vinyl Alcohol (EVOH)

100*

Polyethylene terephthalate (PET)

55*

Linear low density polyethylene (LLDPE)

17*

Low density polyethylene (LDPE)

18**

High-density polyethylene (HDPE)

5-10**

Polypropylene (PP)

19**

Oriented polypropylene (BOPP)

9**

Polyamide (PA)

300**

Oriented polyamide (BOPA)

145**
110-160**

Polystyrene (PS)
Notes:

*Flexible Packaging Solution. Available from:
http://www.dura-pack.com/PDF/Bags-FilmRev1.pdf.
**The barrier performance of common plastic film. Available from:
http://www.evergreen-packaging.com/en/ShowNews.asp?id=70

4.2.6 Oxygen barrier properties of hemicellulose based films
The values of oxygen transmission rate (OTR) and oxygen permeability (OP) of
hemicellulose based films from four different approaches are listed in Table 7, due to the lack of
availability to collect the data from the films made from the other two approaches. These values
were one to two orders of magnitude lower than those reported for hemicellulose based films
(Hartman, 2006; Mikkonen et al., 2009; Mikkonen et al., 2012; Heikkinen et al., 2013;
Heikkinen et al., 2014); also much lower than that of HDPE (600 cm3/(d•m2)), and slightly
lower than that of PET (60 cm3/(d•m2)). Grubb in 2010 reported that a xylan-based barrier had a
76

similar OTR (cc/m2/day) as polyvinylidene chloride (PVdC). Compared with cellophane, a
successful commercialized cellulose product, the hemicellulose films in this study also present
superior oxygen barrier (267.8×10-9 cm3•m / d•m2•KPa, 100% RH in this study versus 3.7×1014 cm3•m / s•m2•Pa, 0% RH) (Tomé et al., 2011). Thus, the hemicellulose films in this study
have superior oxygen barrier properties to most current commercialized packaging polymers.
The previous studies about hemicellulose films mostly investigated the type and amount of
plasticizer and crosslinking agent added to different types of hemicellulose films. Heikkinen et al.
reported that sorbitol plasticized hemicellulose film (40%) showed a higher oxygen permeability
coefficient than that in this study (Heikkinen et al., 2014). For hemicellulose film with increasing
sorbitol amount added, slightly higher OP value was observed (Mikkonen et al., 2009). Study of
the Ara/Xyl ratio of arabinoxylan film suggested that films with larger substitution had higher
OP values as well (Heikkinen et al., 2013). In this study, no significant difference was found
among the different types of hemicellulose films. There were no differences in the type or
amount of the additives in the tested hemicellulose films, and all of the hemicelluloses were
extracted from wood. These are probably the reason why no significant differences were
observed in this study.

77

Table 7 Oxygen permeability of hemicellulose based films from different approaches

Properties
Thickness,
OTR, cm3/(d•m2)
Permeability,10-9
cm3•m/(d•m2•KPa)

CTMP
Aspen
Bleached
61.0 (7.6)
20.7 (19.6)a

CTMP
Aspen
Unbleached
55.9 (7.6)
14.8 (11.2)a

CTMP
Spruce
Bleached
121.9 (10.2)
15.5 (8.7)a

HPP
Bleached
106.7 (20.3)
23.8 ( 16.4)a

HDPE PET
600

60

125.5
85.0
193.6
267.8
(117.7)a
(70.6)a
(93.1)a
(226.9)a
Note: The numbers in parentheses are STDEV.
All measurements were performed in 6 replicates.
Values in the same row with the same superscripts are not significantly different (p 0.05).
HDPE/PET: The barrier performance of common plastic film. Available from:
http://www.evergreen-packaging.com/en/ShowNews.asp?id=70

78

5 FACTORS THAT INFLUENCE THE PROPERTIES OF EXTRACTED
HEMICELLULOSE BASED FILMS
For better understanding the changes in the properties of the hemicellulose films from the
six different approaches, further tests and analyses were conducted.
5.1 Composition of extracted hemicellulose samples
Table 8 shows the compositions of hemicellulose samples obtained from the six different
approaches. The results show that xylan is the main hemicellulose type contained in these
extracted hemicellulose samples, followed by mannan. Meanwhile, small amount of glucan,
galactan and arabinan are contained in these hemicellulose samples, no more than 5% for each of
them. They are attributed to the substitutes in xylan backbone chain. No statistically significant
difference was observed in the amount of xylan obtained between hardwood and softwood. They
are very close. But mannan amounts are somewhat higher in softwood hemicellulose samples
compared with that in hardwood samples. This can be attributable to the difference in the main
sugar unit of hemicelluloses between hardwood and softwood. Softwood hemicellulose is known
to contain more mannan than hardwood hemicellulose (Sjöström, 1993). Klason lignin as
impurity was found in these different hemicellulose samples in small amounts. The percentage of
lignin in the unbleached hemicellulose samples was larger than that in the bleached
hemicellulose samples. No ash or acetyl groups were observed in any of the six samples. The
strong extraction condition (10% potassium hydroxide) was believed to be responsible for the
elimination of acetyl groups (Teleman et al., 2000).

79

Table 8 Composition of hemicellulose samples

Hemicellulose

Glucan
(%)

Xylan
(%)

Mannan
(%)

Galactan
(%)

Arabinan
(%)

Klason lignin
(%)

CTMP
Aspen Bleached

2.0 (0.5)

73.9 (1.4)a

5.1 (2.4)a

2.5 (0.9)

4.0 (1.6)

1.0 (0.1)ab

CTMP
Aspen Unbleached

1.6 (0.7)

73.8 (0.8)a

4.2 (3.0)a

3.7 0.8)

1.8 (1.0)

CTMP
Spruce Bleached

1.9 (1.3)

74.0 (0.6)a

9.7 (2.2 )a

2.8 (1.1)

CTMP
Spruce Unbleached

2.1 (0.2)

73.9 (1.7)a

9.5 (0.8 )a

HPP Bleached

1.6 (0.1)

76.5 (0.6 )a

HPP Unbleached

1.4 (1.2)

75.8 (1.0)a

Ash
(%)

Acetyl
(%)

Total (%)

N/D

N/D

88.6 (1.5)

1.2 (0.3)ab

N/D

N/D

86.4 (1.5)

2.1 (1.9)

0.6 (0.1)a

N/D

N/D

91.1 (1.5)

4.6 (0.2)

1.3 (0.7)

1.1 (0.2)ab

N/D

N/D

92.5 (0.9)

5.6 (2.3)a

3.4 (1.0)

3.0 (0.7)

1.1 (0.3)ab

N/D

N/D

91.2 (1.2)

5.3 (1.8)a

2.8 (0.5)

1.3 (1.6)

1.6 (0.5)b

N/D

N/D

88.2 (1.3)

Notes: Values are reported as mean of two replicates, the numbers in parentheses are the standard deviation.

Values in the same column by the same letter are not significantly different (p 0.05).
N/D =not detected

80

5.2 Molecular weight information analysis of hemicellulose samples
A linear relationship with a R2 of 0.99 was found between the logarithm of the molecular
weight and the volume eluted of the dextrans with known molecular weight used as reference as
illustrated in Figure 18. The linear relation between Log (MW) as Y and the elution volume (V)
is expressed in the equation below:
Y=-0.38V+10.3
Clearly indicates a negative slope (-0.38) suggesting a decrease of Log (MW) with an
increase of V. The smaller the elution volume is, the larger the molecular weight.
The elution profile of hemicellulose extracted from bleached CTMP aspen is shown in
Figure 19 as an example. The other elution profiles of hemicellulose samples are shown in
Figure 28 in the appendix. There are mainly four elution peaks present in the elution profiles of
the extracted hemicellulosic preparations. The peak with elution time from 17 min to 20 min (1st
peak) was assumed to be from the aggregation of hemicellulose, which could be significantly
reduced by using longer sonication time. Similar observations were reported by Stoklosa and
Hodge (Stoklosa and Hodge, 2012) and Skaake et al. (Skaake et al., 2001) with hardwood
glucuronoxylan extracted with KOH. Some researchers also noted that arabinoxylan from rye
bran aggregated to clusters as a function of solvent and time (Ebringerova et al., 1994). They
noticed that these clusters could be partly broken either by exposing the solution to shear force
during intense shaking or by heating. The molecular weight of the peak at elution time from 30
to 32.5 min (3rd peak) was lower than 3000 g/mol and assumed to be from oligosaccharides or
oligomers.
A peak was shown from elution time about 33 min to 34 min, which was followed by a
negative peak. It was believed that this peak was attributed to the encounter with the permeation

81

limit of the column system (Ultrahydrogel 250 1K-80K; Ultrahydrogel 500 10K-400K, Waters),
which was not observed in the profile eluted using a single column (Ultrahydrogel 250). The
figures are shown in Figure 29 in the appendix.
The peak around elution time 35 min (4th peak) is below the permeation limit for the
column system and represents all the other contaminants with much lower molecular weight than
1K contained in these samples, e.g. salts.
The elution peak from elution time about 20 min to 30 min (elution volume 12 ml to 18
ml, 2nd peak) covered the range of molecular weights from 3K to 500K, which was taken as the
peak standing for the primary extracted hemicellulose molecules, separately shown in Figure 20.
The elution profiles of xylans from birchwood and oat spelts freshly made and after 5 days
storage with longer sonication are shown in Figure 30 in the appendix. As is seen in the
chromatograms, the 1st peak presented increased RI signals after 5 days storage compared to the
freshly made samples, yet the 2nd peak kept the same response. This had a good agreement with
aggregation phenomenon. Thus, the second peak was taken as the primary hemicellulose peak
for calculation of molecular weight averages. The same behavior was observed in research by
Saake et al. in 2001.
Therefore, the molecular weight information such as number average molecular weight
(Mn), weight average molecular weight (Mw) and polydispersity index (PDI) of the extracted
hemicellulose samples, was calculated based on this peak (2nd peak).
The molecular weight averages and distributions are shown in Table 9. The molecular
weight averages (Mn and Mw) from the hybrid poplar powder sample were larger than those from
pulp samples in general. This agreed with research by Jacobs and Dahlman did in 2001. They
found that in most cases, the hemicelluloses extracted from wood displayed higher molecular

82

weight than did the corresponding hemicelluloses from chemical pulps. It seemed that the type of
raw materials played an important role on the extracted hemicellulose molecular weight averages.
Bleaching made the hemicellulose molecules from hybrid poplar significantly smaller than the
unbleached ones. For both Mn and Mw, the largest average molecular weight came from the HPP
unbleached hemicellulose sample, 46.6 and 112.8 KDa, respectively. These values were
somewhat higher than the molecular weight values reported for fast-growing poplar wood, maple,
beechwood (Sun et al, 2001; Stoklosa and Hodge, 2012; Saake et al, 2001). This was probably
attributed to the variance of the extraction method (Sun and Hughs, 1998; Peng et al, 2012),
solvent quality (Sakke et al, 2001; Izydorczyk and Biliaderis, 1995) and eluted peak ranges for
calculation (Sun et al, 2001). The polydispersity index (PDI) had similar results to the molecular
weight averages, in the opposite way. Hemicellulose from CTMP aspen bleached treatment
displayed the largest PDI. Polymers with lower PDI usually display better mechanical properties
than those with larger PDI.
The relationship between hemicellulose film properties (TS and MOE, obtained in
chapter 4) and molecular weight / distribution obtained by using SEC two columns system
demonstrate a linear relationship that larger molecular weight favoring larger TS and MOE; the
PDI versus for TS, no significant linear relationship was found between MOE and PDI (Figure
17). This relationship could not be observed from the analysis by the SEC one column system.
Two columns separated the hemicellulose molecules with different molecular weight more
clearly, which improved the accuracy of calculation of the hemicellulose molecular weight.

83

90

Tensile strength (MPa)

80
70
60
50
40
y = 1.0035x - 35.421
R² = 0.6527

30
20

10
0
85

90

95

100

105

110

115

Mw (KDa)
90

Tensile strength (MPa)

80
70
60
50
40

30

y = 1.633x + 2.2944
R² = 0.8396

20
10
0
25

30

35

40

45

50

Mn (KDa)
90
Tensile strength (MPa)

80
70
60
50
40
30

y = -40.26x + 170.35
R² = 0.8204

20
10
0
2.2

2.4

2.6

PDI

2.8

3

3.2

Figure 17 Relationship between films TS/ MOE and molecular weight/ distribution
84

Figure 17 (cont’d)
4
3.5

MOE (GPa)

3
2.5
2
1.5

y = 0.0498x + 0.9749
R² = 0.6502

1
0.5
0
25

30

35

40

45

50

Mn (KDa)
4
3.5

MOE (GPa)

3
2.5
2
1.5

y = 0.0382x - 0.9152
R² = 0.788

1
0.5
0
85

90

95

100
Mw (KDa)

105

110

115

4
3.5

MOE (GPa)

3
2.5
2
y = -0.7885x + 4.9197
R² = 0.2618

1.5
1
0.5
0

2.2

2.4

2.6

2.8
PDI

85

3

3.2

7
2000000

6

100000

Log MW

5
70000

40000

4

6000

3

2
y = -0.3813x + 10.329
R² = 0.9878

1
0
6

8

10

12

14

16

18

20

Elution volume (ml)
Figure 18 Calibration curve of dextrans with known molecular weights

10000

8000

RI detector response

6000

4000

2000

0
0

5

10

15

20

25

30

35

40

45

-2000

-4000

Elution time (min)

Figure 19 SEC elution profile of hemicellulose extracted from bleached CTMP aspen

86

50

3000

RI detector response

2500

2000

1500

1000

500

0
20.5

21.5

22.5

23.5

24.5

25.5

26.5

27.5

28.5

29.5

Elution time (min)

Figure 20 SEC expanded elution profile of hemicellulose from bleached CTMP aspen

87

Table 9 Molecular weight averages of extracted hemicelluloses specimens and their distributions
CTMP A B

CTMP A UB

CTMP S B

CTMP S UB

HPP B

HPP UB

Mn, KDa

28.9 (1.7)a

32.8 (1.2)ab

33.3 (0.8)ab

37.2 (0.8)bc

41.0 (1.7)cd

46.6 (4.5)d

Mw, KDa

90.9 (1.2)a

89.6 (3.5)a

91.4 (2.3)a

94.9 (3.2)a

103.6 (1.6)b

112.8 (2.4)c

PDI

3.15 (0.17)a

2.73 (0.02)bc

2.74 (0.12)b

2.55 (0.03)bc

2.53 (0.07)bc

2.43 (0.17)c

Note:

Values are reported as mean of three replicates, the numbers in parentheses are the standard deviation.
Values in the same row by the same letter are not significantly different (p 0.05).
A aspen, B bleached; S spruce, UB unbleached.

88

5.3 Molecular weight information of purchased xylans & HPPB hemicellulose samples and the
tensile properties of their films
The values of degree of polymerization (DP) for the purchased xylans & bleached
extracted hemicellulose samples from HPP before / after enzymatic hydrolysis and the tensile
properties of their films are summarized in Table 10. The DPs were obtained by two methods,
the reducing-end method (Red. End) and SEC method. For xylan from oat spelts, its ̅̅̅̅̅ from
reducing-end method (130) is much smaller than it is from the SEC method (459), which
suggests xylan from oat spelts has large substitutes in its backbone chain. It is a highly branched
polymer. This leads to its relatively great water solubility, which makes it easier to build
networks among hemicellulose molecules and thus to form film (Mikkonen et al., 2012). The
number average DPs of the three types of xylans estimated from the reducing-end method are
very close. They are also close to the DPs of xylans from beechwood and birchwood obtained by
the SEC method. The hemicellulose sample from HPPB after enzymatic hydrolysis was smaller
than that of the sample without enzymatic hydrolysis. The DPs of xylans obtained by the SEC
method were smaller than the values for HPPB except for the xylan from oat spelts. The PDI
results of xylans from beechwood and birchwood were close to each other and smaller than that
from oat spelts. And also, the enzymatic hydrolysis slightly increased the PDI for the
hemicellulose sample from HPPB.
The TS of xylan film from oat spelts was significantly smaller than that of the xylan films
from birchwood and beechwood (p<0.05). Yet, the elongation was much larger than for those
films. MOE values among the three samples show significant differences from each other.
After enzymatic hydrolysis of bleached HPP hemicellulose sample, the degree of
polymerization, PDI, TS and Elongation all decreased (Table 10). The enzyme cut the

89

hemicellulose molecules shorter, which led to the reducing of TS. This corresponds to the
analysis about the relationship between TS and molecular weight. The result confirms that
molecular weight is an important factor for the tensile properties of hemicellulose film, which is
also suggested by Gröndahl et al (Gröndahl et al., 2004; Heikkinen et al., 2013). Heikkinen et al
also suggested the substitution was a considerable factor for attention.
In addition, the DPs (Red. End) for purchased xylans and DPs (SEC) for extracted HPPB
hemicellulose samples before and after enzymatic hydrolysis present a good linear relationship
with their corresponding tensile strength (Figure 21). This is in good agreement with the study on
the relationship between TS / MOE and molecular weight earlier in this section (Figure 17).
80

Tensile strength (MPa)

70

60
50
40
30

y = 0.2115x + 0.5385
R² = 0.9484

20
10
0
100

150

200

250

300

350

Figure 21 Relationship between the TS of hemicellulose films and hemicellulose DPs
Therefore, molecular weight (for hemicellulose with fewer substitutions, molecular
weight corresponds to its degree of polymerization) is believed to have a significant effect on the
tensile properties of hemicellulose film.

90

Table 10 Comparison of purchased xylans and hemicellulose sample from HPPB and their films
̅̅̅̅̅(Red. End)

̅̅̅̅̅ SEC

̅̅̅̅̅ SEC

PDI

Tensile strength (MPa)

Elongation (%)

MOE (GPa)

Oat spelts

130

459 (13)a

1514 (91)a

3.30 (0.16)a

25.4 (2.7)d

10.5 (1.5)a

1.3 (0.1)d

Beechwood

121

145 (1)d

301 (4)d

2.08 (0.04)d

31.6 (3.1)c

1.9 (0.4)d

2.4 (0.3)b

Birchwood

161

130 (1)d

289 (7)d

2.22 (0.05)cd

30.2 (2.8)c

2.7 (0.5)cd

1.8 (0.2)c

HPP B

━

310 (24)b

768 (59)b

2.48 (0.20)bc

66.8 (3.7)a

7.4 (1.1)b

2.9 (0.2)a

HPP B (hydrolysis)

━

200 (19)c

534 (30)c

2.63 (0.07)b

43.7 (3.6)b

4.1 (0.6)c

2.9 (0.3)a

Xylan sources

Note:

Values are reported as mean of 3 replicates for SEC analysis, 8 replicates for mechanical analysis except that HPP B is 20
replicates.
The numbers in parentheses are the standard deviation.
Values in the same column by the same letter are not significantly different (p 0.05).

91

5.4 Lignin identification and semi-quantification
Fourier transform infrared spectra of all samples are shown in Figure 22. Two different
sources of lignin displayed coincident characteristic peaks, which were in good agreement with
the AlcellTM organosolv lignin study done in 2004 (Boeriu et al., 2004). The occurrence of
obvious bands at 1597, 1514 and 1425 cm-1 involving aromatic skeleton vibrations and the peak
at 1460 cm-1 due to the combination of C-H deformation and aromatic ring vibration can be
observed for all samples, except for xylan from oat spelts. These peaks are the typical
characteristic peaks for lignin (Seca et al., 1998; Boeriu et al., 2004; Peng et al., 2012; Egüés et
al., 2013; Rodrí
guez-Gutiérrez et al., 2014). Their presence confirmed the existence of lignin in
those samples. The xylan from oat spelts were purchased from Sigma-Aldrich had a high
concentration of hemicellulose, and the typical peaks for lignin were absent from its FTIR
spectrum. The unbleached HPP hemicellulose sample exhibited more obvious peaks than other
samples. The absorption bands below 1400 cm-1 are complicated, which can be attributed to
various vibration modes. This makes it difficult to analyze the monolignol unites vibrations in
this region. The peak at 1269 cm-1 shows the vibration for the guaiacyl unit (guaiacyl ring and
C=O stretching) and a vibration at 1213 cm-1 can be attributed to the C-C plus C-O plus C=O
stretching. The most obvious peak at 1043 cm-1 is a complex vibration attributed to the aromatic
C-H deformation associated with the C-O, C-C stretching and C-OH bending in polysaccharides
(Boeriu et al., 2004). The unbleached HPP hemicellulose sample displayed more obvious peaks
on these region compared with the other samples. All of the above indicate that the unbleached
HPP hemicellulose sample contains a relatively higher content of lignin than the bleached HPP
hemicellulose sample. Bleaching is an effective approach for delignification of the alkaline
extracted hemicellulose sample. The peak at 1043 cm-1 is assigned to the C-O, C-C stretching

92

and C-OH bending in xylans and the small band at 896 cm-1 to the characteristic β-glycosidic
bonds between sugar units (Peng et al., 2012). And the shoulder peaks (1160 cm-1 and 990 cm-1)
near the most obvious peak indicate the presence of arabinosyl side chains (Buranov and Mazza,
2010; Peng et al., 2012; Egüés et al., 2013). These results indicate that xylan is the dominant
hemicellulose in the bleached and unbleached HPP samples; arabinosyl units are attached to the
xylan backbone chain. All of the results obtained by this FTIR analysis corresponded to the data
obtained in the HPLC composition analysis (Table 16).

93

(a)

0.8

42-7 bmwel lignin
0.7

Repap alcell lignin
HPP

0.6

Xylan from oat spelts
0.5

Bleached HPP hemicellulose
Unbleached HPP hemicellulose

0.3

0.2

0.1

0
2500

2400

2300

2200

2100

2000

1900

1800

1700

1600

1500

1400 1300 1200 1100 1000

900

800

700

600
-0.1

Wavenumber (cm-1)
-0.2

Figure 22 FTIR spectra of HPP hemicellulose samples, HPP, xylan and lignin (a) and its expanded spectra (b)

94

Absorbance

0.4

Figure 22 (cont’d)

(b)

1597

1514

1460

1425

1113

1269 1213

990
896

Absorbance

1160

1043

1800

1600

1400

1200

Wavernumber (cm-1)

95

1000

800

600

5.5 Crystallinity of hemicellulose based films
Hemicellulose is an amorphous polymer in its native status; but it can form crystalline
structure under some conditions, such as decreasing side group substitution (Höije et al., 2008).
Xylan and sorbitol bought from Sigma-Aldrich, bleached & unbleached extracted hemicelluloses
(BH & UH), and their films without additives (BHF & UHF), with sorbitol (BHSF & UHSF),
and with sorbitol and glutaraldehyde (BHSGF & UHSGF) were scanned by X-ray for structure
analysis with results shown in Figure 23. The additive, sorbitol, presents sharp distinct peaks in
the scanned range 2 from 10ºto 45°. No obvious crystalline peak could be observed from the
X-ray diffraction patterns of purchased xylan and both of the bleached & unbleached extracted
hemicelluloses in the scanned range (Figure 23 a and b). The slight fluctuation observed in both
of the extracted hemicellulose diffractograms was likely due to the scattering from the
amorphous structure. However, the extracted hemicellulose based films BHF / UHF & BHSF /
UHSF & BHSGF / UHSGF (without additives & with sorbitol & with sorbitol and
glutaraldehyde) all exhibited broad but obvious peaks (corresponding to reflections 100, 101 and
011, 110, 202 and 022) (Heikkinen et al., 2013), even though BHF &UHF films do not contain
the crystalline sorbitol. Within this test condition, no sorbitol characteristic peaks were observed
in the diffractograms of these films. These results suggest that hemicellulose possesses some
crystalline structures in the formed film. This result is in agreement with the other studies on
different types of hemicelluloses (different types of xylans and mannan) (Gröndahl et al., 2004;
Peura et al., 2008; Mikkonen et al. in 2009; Zhang et al., 2011; Heikkinen et al., 2013; Heikkinen
et al., 2014). Similar crystallization behavior was also found in the study of starch (amylopectin)
films (Anglès and Dufresne, 2000). It was reported by Peura et al that during film forming (water
evaporation), glycerol plasticized oat spelts xylan formed new crystallite nucleations resulting in

96

increasing crystallinity, but the average size of the crystallites remained constant (Peura et al.,
2008). Gröndahl et al. also stated that sorbitol and xylitol plasticized aspen glucuronoxylan films
displayed a crystalline structure, and the increasing amount of plasticizer added in those films
did not increase its crystallinity, but increased the distance between the lattice planes making the
peak positions move to lower angles under the tested condition that the amount of plasticizer
added was no more than 50% (Gröndahl et al., 2004). Similar to aspen xylan, this study displays
the same shift direction following the addition of the additives. Gröndahl et al. also suggested
that sorbitol could interact and even cocrystallize with xylan leading to the increasing of the
spacing between the crystal lattice planes, but Mikkonen et al. believed that it was caused by
different degrees of hydration (Mikkonen et al. in 2009). But both studies suggested that
plasticizer may increase the mobility of the polymer chains, therefore promoting crystallization
of the xylan molecules. Substitution was suggested to prevent crystallization. Lower Ara/Xyl
ratio (0.14, 0.23, 0.27 compared with 0.34, 0.56) and lower β-glucan/AX (2%, 6% compared
with 13%, 18%) were reported to enable the crystallization of xylan molecules; no crystal
structure was detected in the films with higher substitution ratios (Mikkonen et al., 2009; Zhang
et al., 2011; Mikkonen et al. in 2012; Heikkinen et al., 2013). Unsubstituted regions along the
xylan chain were reported to promote crystallization (Dervilly-pinel et al., 2004). From the
results of the composition analysis portion of this study, no acytel groups were detected in the
extracted hemicellulose samples in this study. This may be the reason for extracted hemicellulose
films possessing such high crystallinity. The average crystallinity of bleached hemicellulose
films with sorbitol and glutaraldehyde was computed as 52.3% (±0.7); and 63.4% (±1.1) for
the unbleached hemicellulose film with both additives by the XRD analysis.

97

(a)

Intensity

Sorbitol

BHSGF
BHSF
BHF
BH
Xylan
10

15

20

25

30

35

40

45

2 Theta

(b)

Intensity

Sorbitol

UHSGF
UHSF
UHF
UH
Xylan
10

15

20

25

30

35

40

45

2 Theta

Figure 23 X-ray diffraction patterns of hemicelluloses and hemicellulose films, (a) bleached; (b)
unbleached (The signal of sorbitol is reduced by a factor of 2.5 times).

98

The X-ray diffraction spectra of xylan films from beechwood, birchwood and oat spelts
are displayed in Figure 24. The crystallinity for xylan films from beechwood, birchwood and oat
spelts are 69.0% (±1.6%), 69.7% (±0.8%) and 62.4% (±1.2%), respectively. The crystallinity
of xylan films from hardwood are significantly larger than that of xylan film from oat spelts
(p<0.05). The presence of the large substitutions of xylan from oat spelts is unfavorable for
hemicellulose molecule to arrange in order for formation of crystalline. They can increase the
spacing between hemicellulose molecules and thus decrease the interaction between them.
Therefore the hemicellulose molecules can relatively move more easily. It is reasonable that the
TS of xylan film from oat spelts is smaller than that from birchwood and beechwood. But
considering the similarity of ̅̅̅̅̅ (Red. End) between xylan from oat spelts and xylan from
hardwood, the difference of TS for xylan films from both of them is not remarkable, but still

Intensity

significant (p<0.05).

10

15

20

25

30

35

40

45

2 Theta
Figure 24 X-ray diffraction patterns of xylan films, from beechwood, birchwood and oat spelts.

99

5.6 Influence of additives on the tensile properties of hemicellulose films
The tensile properties of both bleached and unbleached HPP hemicellulose made films
(HPPB & HPPU) and with sorbitol (HPPBS &HPPUS), with sorbitol and glutaraldehyde
(HPPBSG & HPPUSG) are listed in Table 11 and Table 12. The functions of sorbitol and
glutaraldehyde were confirmed by this test. As is seen in Table 11, the addition of sorbitol
significantly decreased the film tensile strength (from 89.6 MPa to 53.1 MPa) and MOE (from
4.9 GPa to 3.0 GPa), but increased the elongation at break (from 3.1% to 4.7%). It can promote
the hemicellulose molecule chain mobility, thus make the film more flexible. This benefits its
application as packaging material. Similar results were observed in the unbleached HPP
hemicellulose samples (Table 12). The addition of glutaraldehyde to the plasticized
hemicellulose film significantly increased the film tensile strength (from 53.1 MPa to 66.8 MPa)
and elongation (from 4.7% to 7.4%), while MOE was not significantly affected (from 3.0 GPa to
2.9 GPa). But the MOE value for the unbleached sample significantly increased after adding
glutaraldehyde (from 3.0 GPa to 3.5 GPa). This corresponded to the results from sorbitolglyoxal-galactoglucomannan film in general (Mikkonen et al., 2012). It was suggested that
glyoxal could interact with the hydroxyl groups on the hemicellulose molecule chains to form
hemiacetal bonds resulting in a network of crosslinked hemicellulose, besides hydrogen bonding
of hemicellulose molecules (Zhao and Deng, 2006; Mikkonen et al., 2012). This is probably the
reason why the tensile strength of unbleached HPP hemicellulose film in this study was superior
to those for similar hemicellulose based films.

100

Table 11 Mechanical properties of HPP bleached hemicellulose films
HPPB

HPPBS

HPPBSG

Tensile strength, MPa
89.6 (4.9)a
53.1 (4.3)c
66.8 (3.7)b
Elongation, %
3.1 (0.4)c
4.7 (1.0)b
7.4 (1.1)a
MOE, GPa
4.9 (0.4)a
3.0 (0.2)b
2.9 (0.2)b
MC, %
14.5 (0.3)
15.3 (0.3)
15.9 (0.7)
Thickness, μm
27.9 (5.1)
33.0 (2.5)
91.4 (10.2)
Breaking length, km
7.6
4.2
5.7
Note: HPP: hybrid poplar powder.
HPPB: HPP bleached hemicellulose made film without additives.
HPPBS: HPP bleached hemicellulose made film with sorbitol.
HPPBSG: HPP bleached hemicellulose made film with sorbitol and glutaraldehyde.
The numbers in parentheses are STDEV.
All measurements were performed in 6 replicates, except for BHPPSG, 20 replicates.
Values in the same row by the same letter are not significantly different (p 0.05).
Table 12 Mechanical properties of HPP unbleached hemicellulose films

HPPU
98.8 (8.1)a
2.6 (0.5)c
5.1 (0.5)a
14.9 (0.5)
30.5 (5.1)
8.4

HPPUS
61.0 (7.1)c
4.0 (0.7)b
3.0 (0.3)c
15.5 (0.2)
38.1 (5.1)
4.8

HPPUSG
76.5 (7.0)b
5.4 (0.9)a
3.5 (0.4)b
17.0 (0.3)
68.6 (5.1)
6.0

Tensile strength, MPa
Elongation, %
MOE, GPa
MC, %
Thickness, μm
Breaking length, km
Note: HPP: hybrid poplar powder.
HPPU: HPP unbleached hemicellulose made film without additives.
HPPUS: HPP unbleached hemicellulose made film with sorbitol.
HPPUSG: HPP unbleached hemicellulose made film with sorbitol and glutaraldehyde.
The numbers in parentheses are STDEV.
All measurements were performed in 6 replicates, except for HHPPSG, 20 replicates.
Values in the same row by the same letter are not significantly different (p 0.05).

101

5.7 Discussion
From the results of the statistical analysis, bleaching was an effective treatment in this
study, responsible for the changing of tensile properties of hemicellulose films from those six
approaches. It could cause the difference of hemicellulose samples, such as hemicellulose
molecular weight, its distribution, structure (side groups and side chain) and also the sample
composition. From the analyses results earlier in this section, it is confirmed that bleaching
caused a reduction in the hemicellulose molecular weight, decreased lignin content and
crystallinity of hemicellulose films, and changed the composition of hemicellulose sample. From
molecular weight analysis, it was shown that the unbleached samples had larger molecular
weights in general, which contributed to the higher tensile properties (TS and MOE).
Furthermore, XRD analysis showed that the unbleached HPP sample had higher crystallinity. In
general, the higher crystallinity a film possesses, the higher tensile strength it exhibits. As is
known, hydrogen bonding between hydroxyl groups in the hemicellulose molecular chains is the
primary interaction in the hemicellulose films. The different hemicellulose molecular weight and
structure could lead to different hydrogen bonding intensity and thus result in differences in the
final tensile strength.
Bleaching was not the only important influence on the tensile properties of the
hemicellulose based films. The sources, like hardwood and softwood; processes (fiber obtained
through the chemical, mechanical or thermal processes or by grinding) also influenced the tensile
properties. Films made from hemicellulose obtained from hardwood CTMP fibers had less
strength than those made from hemicellulose obtained from softwood CTMP fibers, but higher
elongation. As seen from the result of composition analysis, hardwood contains less mannan.
The monosaccharide composition is different. Mannan is composed of hexose sugars, while

102

xylan is composed of pentose sugars. The number of hydroxyl groups will be higher in mannan
than in xylan. It is possible this leads to more hydrogen bonding between mannan molecules and
additives, resulting in a more compact film structure. This corresponds to the results of the study
on

glycerol

and

sorbitol

plasticized

oat

spelt

arabinoxylan

(OsAX)

and

spruce

galactoglucomannan (GGM) (Heikkinen et al., 2014). It might be also because hardwood fiber is
more sensitive to the chemical, mechanical and thermal treatment during the pulping process. If
raw materials are processed too much, this can have a great impact on the tensile properties of
the films made from it. This could significantly reduce nearly all the tensile properties of the
hemicellulose based films, including tensile strength, elongation and MOE, regardless of the
bleaching procedure, as is shown for films from aspen CTMP fiber and hybrid poplar powder.
This is suggested to be due to the relatively high degradation of hemicellulose molecules during
those chemithermomechanical pulping procedures, which resulted in a lower molecular weight in
general.
And also, the presence of lignin may contribute to the promotion of the tensile strength,
since it could be involved in the interaction with different molecules in the hemicellulose based
films by forming chemical bonds or other physical bonds. In plant cell walls, lignin crosslinks
hemicellulose by covalent bonds (mainly R-benzyl ether linkages). This is believed to promote
hemicellulose film formation (Hansen et al., 2012). It was also reported that without sufficient
lignin addition, birchwood xylan could not form a continuous film alone (Goksu et al., 2007;
Hansen et al., 2012). Therefore, it is suggested that lignin helps enhance the interaction among
the hemicellulose molecules in the films, decrease the potential relative slippage between
hemicellulose molecules, and then improve the tensile strength of these films, thus reducing the

103

percentage elongation at break. This corresponds to the results of earlier analyses about lignin
content and tensile properties of hemicellulose films.
In all of above, the mechanical properties obtained in this study were probably attributed
to the hemicellulose source and isolation method with proper plasticizer and crosslinking agent
addition. The better performance of the HPP unbleached sample is likely attributable to the
larger average molecular weight, higher lignin content and higher crystallinity. These
characteristics had a good agreement with the mechanical and barrier properties in general.
The tensile strength and modulus of elasticity of hemicellulose films in this study are
higher than those of the commonly used packaging plastics. But the values of elongation at break
are much smaller, except for some types of polystyrene (PS) and polyvinyl chloride (PVC). The
hemicellulose films also present superior oxygen barrier property, even better than PET.
However, they still present poor water vapor barrier compared with the commonly used
packaging petro-plastics. These features lead hemicellulose products to packaging applications
with relatively high tensile strength and oxygen barrier requirements but a low humidity
environment. But actually, in reality it is relatively hard to avoid the impact of moisture in the
environment. One solution is to coat on or to laminate with plastics / paper; the composite can
provide sufficient barrier and mechanical properties for those food packaging applications. Thus,
it will be suitable to be used in food packaging applications such as chip bags, instant coffee
boxes, beverage boxes, and nut bags. Another solution is just to design these as water soluble
packages, such as the water soluble seasoning sachet in an instant noodle bag or certain candy
inner layers in packages. A hemicellulose based barrier application has been already been
successfully used as a large spice bag, waiting to be dispersed in industrial scale cooking
operations (Grubb, 2010).

104

6 CONCLUSION
Hemicellulose obtained through alkaline extraction and ethanol precipitation procedures
in this study were mainly composed of xylan, slightly contaminated by lignin. Slightly higher
levels of lignin were present in the unbleached samples compared with the bleached ones. No Oacetyl groups were detected in the hemicellulose samples. With the approaches used in this study,
a free-standing hemicellulose based film could be obtained with good transparency, a smooth
surface, and a relatively uniform structure.
Powdered hemicellulose samples possessed higher molecular weight than the pulp
samples; Molecular weights of unbleached samples were higher than the bleached samples, in
general. The polydispersity index behaved in the opposite manner. The FTIR analysis confirmed
that the unbleached hemicellulose samples contained more lignin. XRD analysis and DSC
analysis proved that crystal regions were formed in the hemicellulose films, which benefited
their propertiess.
These hemicellulose based films possessed superior strength. Among these six different
approaches, hemicellulose from unbleached HPP had the highest tensile strength at 76.5 MPa,
with specific strength 58.8×103 N·m/kg and MOE 3.5 GPa. This is superior to the tensile
strength from other hemicellulose based films and common commercial plastics, which are
usually below 50 MPa; and comparable or higher MOE (from 0.25 GPA to 3.4 GPa).
Glutaraldehyde appeared to enhance the film properties significantly and made the tensile
strength in this study superior to the other hemicellulose films. All the unbleached samples
displayed larger strength than the bleached samples. These differences were statistically
significant. MOE was slightly larger for the unbleached samples compared with the bleached
ones, except for aspen CTMP. The elongation at break was affected oppositely. Molecular

105

weight and lignin content were suggested to have significant effects on the tensile properties.
Combined with crystallinity, this resulted in the differences in the tensile properties for the
hemicellulose films from different approaches.
Comparable water vapor and oxygen barrier property to results from similar
hemicellulose studies were obtained in this study. Better water vapor barrier properties were
obtained for the unbleached hemicellulose based films, but they were still several orders
magnitude lower than for the polymers currently used in packaging.
In all of above, the source of hemicellulose and its pretreatment had a great influence on
the physical and mechanical properties of hemicellulose based films. These influences could
provide useful information for guidance in the utilization of hemicellulose on an industrial scale.

106

7 FUTURE WORK
The hemicellulose source is suggested to be an important role on its film properties in
this study. However, the reasons for the effects on properties are not well known. Molecular
weight of the hemicellulose is proposed in this study to have a linear relationship with the tensile
strength of the film. The hemicellulose type was not thoroughly investigated for its influence on
the film properties. Most of the research focused on the properties of one type of hemicellulose
(e.g. xylan) film, even though this type of hemicellulose may be from different sources. Few
studies could be found in the open literature comparing film properties from different types of
hemicelluloses. Only assumptions for the different behavior were suggested in the study of
plasticized OsAX and GGM films (Heikkinen et al., 2014). Thus, more investigation needs to be
conducted to determine the impact of hemicellulose type on the film properties, which would
benefit the potential utilization of hemicellulose film as a packaging material on an industrial
scale.
Another rising problem for hemicellulose film for applications in packaging is its longterm stability. Every packed item has a shelf life. During the shelf life, the physical stability of
the hemicellulose film is crucial in determining if enough protection will be provided for the
item, especially if small molecules (such as plasticizers and cross-linking agents) are added to
the hemicellulose films. A study of the migration of plasticizers and crosslinking agents would
give significant guidance in determining the stability of hemicellulose films. Some differences in
film properties were observed for glycerol and sorbitol plasticized OsAX and GGM films after
storage for 4 months (Heikkinen et al., 2014).
As a bio-renewable and biodegradable polymer, hemicellulose film is counted on as a
good sustainable film for packaging applications. However, most of the research was conducted

107

on the biodegradation of PLA, PHA, starch, cellulose, chitosan and protein products (Rudnik,
2013 b). The biodegradability of hemicellulose products is seldom investigated, especially after
plasticizers and cross-linking agents are added. Testing the biodegradability of hemicellulose
products would be very helpful for the guidance of hemicellulose application as sustainable
packaging materials.
Besides its biodegradability, hemicellulose has another well

known feature,

hygroscopicity, because it is very rich in hydroxyl groups. This feature makes the properties of
hemicellulose products particularly sensitive to the presence of moisture in its surrounding
environment. Many studies reported that the humidity gradient could significantly impact the
permeability of water vapor for hydrophilic films. The higher the relative humidity is, the larger
the WVTR value (Hagenmaier and Shaw, 1990; McHugh et al., 1993; Mikkonen et al., 2009).
And also, it was reported that the strength of flax fiber increased first and then decreased
following an increase in the relative humidity from 21% to 85% (Thuault et al., 2015). However,
few reports of the impact of relative humidity on the mechanical properties of hemicellulose
products could be found in the open literatures. This information would be very useful for
guidance on hemicellulose used as a packaging material in different environments.
In addition, the solvent casting method in this study limits hemicellulose film potential
for manufacturing on an industrial scale. Exploration of fabrication by extrusion or injection
molding is recommended; as such traditional polymer processing techniques would fit the
requirements for industrial manufacture. So far, starch, cellulose, PLA and PHA based products
are widely investigated and even commercially produced (Plackett, 2011; Ebnesajjad, 2013).

108

APPENDIX

109

Table 13 Correlations of tensile strength, elongation and MOE
TS
TS

Pearson Correlation

El

1

.187

Sig. (2-tailed)
N
El

MOE

129
*

Pearson Correlation

.187

Sig. (2-tailed)

.034

N

129

Pearson Correlation

.581

MOE
*

.581**

.034

.000

129

129

1

-.200*
.023

129

**

-.200

Sig. (2-tailed)

.000

.023

N

129

129

129
*

1

129

*. Correlation is significant at the 0.05 level (2-tailed).
**. Correlation is significant at the 0.01 level (2-tailed).

Table 14 Multivariate Tests for assessing the impact of treatment on tensile properties
Effect
Intercept

Treatment

Value

F

Hypothesis df

Error df

Sig.

b

3.000

120.000

.000

Pillai's Trace

.985

2661.502

Wilks' Lambda

.015

2661.502b

3.000

120.000

.000

Hotelling's Trace

66.538

2661.502

b

3.000

120.000

.000

Roy's Largest Root

66.538

2661.502b

3.000

120.000

.000

1.782

29.730

18.000

366.000

.000

.044

38.144

18.000

339.897

.000

7.061

46.553

18.000

356.000

.000

5.149

c

6.000

122.000

.000

Pillai's Trace
Wilks' Lambda
Hotelling's Trace
Roy's Largest Root

104.690

a. Design: Intercept + Treatment
b. Exact statistic
c. The statistic is an upper bound on F that yields a lower bound on the significance level.

110

Table 15 Univariate tests for assessing the impact of treatment on tensile properties
Dependent
Source

Variable

Type III Sum
of Squares

df

Mean Square

F

Sig.

a

5

2632.941

100.301

.000

143.388b

5

28.678

21.172

.000

c

5

2.720

34.955

.000

TS

459782.477

1

459782.477 17515.282

.000

El

4342.949

1

937.204

1

TS

13164.705

5

2632.941

100.301

.000

El

143.388

5

28.678

21.172

.000

13.602

5

2.720

34.955

.000

TS

3228.794

123

26.250

El

166.601

123

1.354

9.573

123

.078

TS

496845.019

129

El

5157.422

129

MOE

1033.759

129

Corrected Total TS

16393.500

128

El

309.990

128

23.175

128

Corrected

TS

Model

El
MOE

Intercept

MOE
Treatment

MOE
Error

MOE
Total

MOE

13164.705

13.602

a. R Squared = .803 (Adjusted R Squared = .795)
b. R Squared = .463 (Adjusted R Squared = .441)
c. R Squared = .587 (Adjusted R Squared = .570)

111

4342.949

3206.351

.000

937.204 12042.074

.000

Table 16 Multiple Comparisons among different treatment levels by Tukey test
Dependent (I)
Variable
TS

Treatment Treatment
T1

T2

T3

T4

T5

T6

95% Confidence Interval

(J)
Mean Difference (I-J)

Std. Error

Sig.

Lower Bound

Upper Bound

1.49502

.020

-9.1471

-.4904

T2

-4.8188

*

T3

-12.5991

*

1.55122

.000

-17.0901

-8.1080

T4

-24.0019

*

1.76438

.000

-29.1100

-18.8938

T5

-19.7920

*

1.38351

.000

-23.7975

-15.7865

T6

-29.5465

*

1.62416

.000

-34.2487

-24.8443

T1

4.8188

*

1.49502

.020

.4904

9.1471

T3

-7.7803

*

1.56647

.000

-12.3155

-3.2451

T4

-19.1832

*

1.77780

.000

-24.3302

-14.0361

T5

-14.9732

*

1.40059

.000

-19.0282

-10.9183

T6

-24.7278

*

1.63874

.000

-29.4722

-19.9834

T1

12.5991

*

1.55122

.000

8.1080

17.0901

T2

7.7803

*

1.56647

.000

3.2451

12.3155

T4

-11.4028

*

1.82532

.000

-16.6874

-6.1183

T5

-7.1929

*

1.46043

.000

-11.4211

-2.9648

T6

-16.9475

*

1.69016

.000

-21.8408

-12.0542

T1

24.0019

*

1.76438

.000

18.8938

29.1100

T2

19.1832

*

1.77780

.000

14.0361

24.3302

T3

11.4028

*

1.82532

.000

6.1183

16.6874

T5

4.2099

1.68511

.133

-.6687

9.0885

T6

-5.5446

*

1.88770

.045

-11.0098

-.0795

T1

19.7920

*

1.38351

.000

15.7865

23.7975

T2

14.9732

*

1.40059

.000

10.9183

19.0282

T3

7.1929

*

1.46043

.000

2.9648

11.4211

T4

-4.2099

1.68511

.133

-9.0885

.6687

T6

-9.7545

*

1.53768

.000

-14.2064

-5.3027

T1

29.5465

*

1.62416

.000

24.8443

34.2487

T2

24.7278

*

1.63874

.000

19.9834

29.4722

T3

16.9475

*

1.69016

.000

12.0542

21.8408

T4

5.5446

*

1.88770

.045

.0795

11.0098

T5

9.7545

*

1.53768

.000

5.3027

14.2064

112

Table 16 (cont’d)
El

T1

T2

T2

1.4348

*

.33960

.001

.4516

2.4180

T3

-1.0162

.35236

.052

-2.0363

.0040

T4

-.3439

.40078

.956

-1.5042

.8164

T5

-1.5869

*

.31427

.000

-2.4967

-.6770

T6

.4200

.36893

.864

-.6481

1.4882

T1

-1.4348

*

.33960

.001

-2.4180

-.4516

T3

-2.4510

*

.35583

.000

-3.4812

-1.4208

-1.7787

*

.40383

.000

-2.9479

-.6095

T5

-3.0217

*

.31815

.000

-3.9428

-2.1006

T6

-1.0148

.37224

.077

-2.0925

.0629

T1

1.0162

.35236

.052

-.0040

2.0363

T2

2.4510

*

.35583

.000

1.4208

3.4812

T4

.6723

.41463

.586

-.5281

1.8727

T5

-.5707

.33174

.521

-1.5312

.3897

T6

1.4362

*

.38393

.004

.3247

2.5477

T1

.3439

.40078

.956

-.8164

1.5042

T2

1.7787

*

.40383

.000

.6095

2.9479

T3

-.6723

.41463

.586

-1.8727

.5281

T5

-1.2430

*

.38278

.018

-2.3512

-.1348

T6

.7639

.42880

.481

-.4775

2.0054

T1

1.5869

*

.31427

.000

.6770

2.4967

T2

3.0217

*

.31815

.000

2.1006

3.9428

T3

.5707

.33174

.521

-.3897

1.5312

T4

1.2430

*

.38278

.018

.1348

2.3512

T6

2.0069

*

.34929

.000

.9957

3.0182

T1

-.4200

.36893

.864

-1.4882

.6481

T2

1.0148

.37224

.077

-.0629

2.0925

T3

-1.4362

*

.38393

.004

-2.5477

-.3247

T4

-.7639

.42880

.481

-2.0054

.4775

T5

*

.34929

.000

-3.0182

-.9957

T4

T3

T4

T5

T6

-2.0069

113

Table 16 (cont’d)
MOE

T1

T2

T3

T4

T5

T6

T2

.0467

.08140

.993

-.1889

.2824

T3

.4091

*

.08446

.000

.1646

.6536

T4

.0501

.09607

.995

-.2281

.3282

T5

-.1641

.07533

.256

-.3822

.0540

T6

-.7609

*

.08844

.000

-1.0169

-.5048

T1

-.0467

.08140

.993

-.2824

.1889

T3

.3624

*

.08529

.001

.1154

.6093

T4

.0033

.09680

1.000

-.2769

.2836

T5

-.2108

.07626

.070

-.4316

.0100

T6

-.8076

*

.08923

.000

-1.0660

-.5493

T1

-.4091

*

.08446

.000

-.6536

-.1646

T2

-.3624

*

.08529

.001

-.6093

-.1154

T4

-.3591

*

.09939

.006

-.6468

-.0713

T5

-.5732

*

.07952

.000

-.8034

-.3430

T6

-1.1700

*

.09203

.000

-1.4364

-.9035

T1

-.0501

.09607

.995

-.3282

.2281

T2

-.0033

.09680

1.000

-.2836

.2769

T3

.3591

*

.09939

.006

.0713

.6468

T5

-.2141

.09175

.189

-.4798

.0515

T6

-.8109

*

.10279

.000

-1.1085

-.5133

T1

.1641

.07533

.256

-.0540

.3822

T2

.2108

.07626

.070

-.0100

.4316

T3

.5732

*

.07952

.000

.3430

.8034

T4

.2141

.09175

.189

-.0515

.4798

T6

-.5968

*

.08373

.000

-.8392

-.3544

T1

.7609

*

.08844

.000

.5048

1.0169

T2

.8076

*

.08923

.000

.5493

1.0660

T3

1.1700

*

.09203

.000

.9035

1.4364

.10279

.000

.5133

1.1085

.08373

.000

.3544

.8392

T4

.8109

*

T5

.5968

*

Based on observed means.
The error term is Mean Square(Error) = .078.
*. The mean difference is significant at the .05 level.

114

Table 17 Multivariate Tests
Effect
Intercept

Value
Pillai's Trace
Wilks' Lambda
Hotelling's Trace
Roy's Largest Root

Bl

Pillai's Trace
Wilks' Lambda
Hotelling's Trace
Roy's Largest Root

PP

Pillai's Trace
Wilks' Lambda
Hotelling's Trace
Roy's Largest Root

Bl * PP

Pillai's Trace

Hypothesis df

Error df

Sig.

7164.266

c

3.000

90.000

.000

.004

7164.266

c

3.000

90.000

.000

238.809

7164.266

c

3.000

90.000

.000

238.809

7164.266

c

3.000

90.000

.000

.535

34.528

c

3.000

90.000

.000

34.528

c

3.000

90.000

.000

34.528

c

3.000

90.000

.000

1.151

34.528

c

3.000

90.000

.000

.872

203.659

c

3.000

90.000

.000

203.659

c

3.000

90.000

.000

203.659

c

3.000

90.000

.000

6.789

203.659

c

3.000

90.000

.000

.292

12.343

c

3.000

90.000

.000

3.000

90.000

.000

.996

.465
1.151

.128
6.789

Wilks' Lambda

.708

12.343

c

Hotelling's Trace

.411

12.343

c

3.000

90.000

.000

12.343

c

3.000

90.000

.000

Roy's Largest Root
a. Wood =

F

.411

Hardwood

b. Design: Intercept + Bl + PP + Bl * PP
c. Exact statistic

115

Table 18 Tests of Between-Subjects Effects
Dependent
ce

Variable

Sig.

Type III Sum of
Squares
b

df

Mean Square

F

3

3908.674

179.388

.000

3

43.251

34.629

.000

3

2.568

36.619

.000

Corrected

TS

11726.021

Model

El

129.753

MOE

7.704

TS

334596.300

1

334596.300

15356.232

.000

El

3059.486

1

3059.486

2449.619

.000

MOE

796.990

1

796.990

11364.851

.000

TS

1212.092

1

1212.092

55.629

.000

El

67.604

1

67.604

54.128

.000

MOE

1.727

1

1.727

24.623

.000

TS

11311.654

1

11311.654

519.146

.000

El

38.630

1

38.630

30.929

.000

MOE

5.389

1

5.389

76.840

.000

TS

139.038

1

139.038

6.381

.013

El

1.868

1

1.868

1.496

.224

MOE

2.364

1

2.364

33.705

.000

TS

2004.584

92

21.789

El

114.905

92

1.249

MOE

6.452

92

.070

TS

359071.065

96

El

3661.779

96

MOE

828.765

96

Corrected

TS

13730.605

95

Total

El

244.657

95

MOE

14.156

95

Intercept

Bl

PP

Bl * PP

Error

Total

a. Wood =

c

d

Hardwood

b. R Squared = .854 (Adjusted R Squared = .849)
c. R Squared = .530 (Adjusted R Squared = .515)
d. R Squared = .544 (Adjusted R Squared = .529)

116

Table 19 Multivariate Tests
Effect
Intercept

Value
Pillai's Trace

Hotelling's Trace
Roy's Largest Root
Bl

Pillai's Trace
Wilks' Lambda
Hotelling's Trace

Wood

Error df

Sig.

c

3.000

74.000

.000

.005

4489.992

c

3.000

74.000

.000

182.027

4489.992

c

3.000

74.000

.000

182.027

4489.992

c

3.000

74.000

.000

.457

20.720

c

3.000

74.000

.000

20.720

c

3.000

74.000

.000

20.720

c

3.000

74.000

.000

3.000

74.000

.000

.543
.840

Roy's Largest Root

.840

20.720

c

Pillai's Trace

.829

119.532

c

3.000

74.000

.000

119.532

c

3.000

74.000

.000

119.532

c

3.000

74.000

.000

4.846

119.532

c

3.000

74.000

.000

.169

5.016

c

3.000

74.000

.003

3.000

74.000

.003

Wilks' Lambda
Hotelling's Trace
Roy's Largest Root
Bl * Wood

Hypothesis df

4489.992

.995

Wilks' Lambda

F

Pillai's Trace

.171
4.846

Wilks' Lambda

.831

5.016

c

Hotelling's Trace

.203

5.016

c

3.000

74.000

.003

5.016

c

3.000

74.000

.003

Roy's Largest Root
a. Raw material format=

.203

CTMP

b. Design: Intercept + Bl + Wood + Bl * Wood
c. Exact statistic

117

Table 20 Tests of Between-Subjects Effects
Dependent
Source

Type III Sum of

Variable

Squares

df

Mean Square

F

Sig.

Corrected

TS

5505.417

Model

El

3

1835.139

68.758

.000

68.697

3

22.899

14.784

.000

MOE

2.203

TS

3

.734

8.247

.000

El

248134.431

1

248134.431

9297.032

.000

2570.228

1

2570.228

1659.358

.000

MOE

513.280

1

513.280

5763.902

.000

TS

1240.829

1

1240.829

46.491

.000

El

20.936

1

20.936

13.516

.000

MOE

.460

1

.460

5.165

.026

TS

4763.138

1

4763.138

178.464

.000

El

36.834

1

36.834

23.780

.000

MOE

.802

1

.802

9.006

.004

TS

204.416

1

204.416

7.659

.007

El

2.742

1

2.742

1.770

.187

MOE

.777

1

.777

8.720

.004

TS

2028.413

76

26.690

El

117.719

76

1.549

MOE

6.768

76

.089

TS

253322.949

80

El

2833.946

80

MOE

554.320

80

Corrected

TS

7533.829

79

Total

El

186.416

79

MOE

8.971

79

Intercept

Bl

Wood

Bl * Wood

Error

Total

a. Raw material format=

b

c

d

CTMP

b. R Squared = .731 (Adjusted R Squared = .720)
c. R Squared = .369 (Adjusted R Squared = .344)
d. R Squared = .246 (Adjusted R Squared = .216)

118

S: G 3:1

S: G 2:1

S: G 1:0

Figure 25 Xylan film from beechwood with different ratios of sorbitol and glutaraldehyde

119

Figure 25 (cont’d)
S: G 1:1

S: G 0:0

S: G 0:1

120

Figure 25 (cont’d)
S: G 1:2

S: G 1:3

121

S/G 0g/0g

S/ G 0.1g/0.025g

S/ G 0.2g/ 0.05g

Figure 26 xylan film from beechwood with different amount of additives

122

Figure 26 (cont’d)
S/ G 0.4g/ 0.1g

S/ G 0.6g/ 0.15g

123

Figure 27 QQ Plots of TS, MOE and El from different treatments

124

Figure 27 (cont’d)

125

Figure 27 (cont’d)

126

Figure 27 (cont’d)

127

Figure 27 (cont’d)

128

Figure 27 (cont’d)

129

Figure 27 (cont’d)

130

CTMP A B 3

6000
5000
4000
3000
2000
1000
0
-1000 0
-2000
-3000

10

20

30

40

50

Elution time (min)

2000
0
0

CTMP A UB 2
5000

4000

4000

3000
2000
1000
0
-1000 0

10

20

30

40

50

-2000
-3000

CTMP A UB 3

3000
2000
1000
0
-1000 0

20

30

40

50

Elution time (min)

CTMP S B 1

5000

14000

4000

12000

3000
2000
1000
0
5

10 15 20 25 30 35 40 45 50

-2000

-3000

10

-2000
-3000

Elution time (min)

0

10 15 20 25 30 35 40 45 50

Elution volume (ml)

5000

-1000

5

-2000

RI detector response

RI detector response

4000

-4000

CTMP A UB 1

RI detector response

RI detector response

6000

RI detector response

RI detector response

CTMP A B 2

10000
8000
6000

4000
2000
0
-2000 0
-4000

Elution time (ml)

5 10 15 20 25 30 35 40 45 50

Elution time (min)

Figure 28 Size exclusion chromatography whole elution profiles from two columns system of
hemicelluloses from different approaches, (A: aspen; B: bleached; S: spruce; UB: unbleached).

131

Figure 28 (cont’d)

14000

14000

12000

12000

10000

10000

RI detector response

CTMP S B 3

8000
6000
4000
2000
0
-2000 0

5

-4000

10 15 20 25 30 35 40 45 50

2000
0

4000

4000

3000

3000

RI detector response

RI detector response

4000

2000
1000

0
-1000

0

5

10 15 20 25 30 35 40 45 50

-2000

40

50

1000
0
-1000

0

5 10 15 20 25 30 35 40 45 50

-2000

Elution time (min)

HPP B 1
5000

3000

4000

2000
1000
0
5

10 15 20 25 30 35 40 45 50

-2000
-3000

30

2000

4000

0

20

Elution time (min)

-3000

Elution time (min)

CTMP S UB 3

-1000

10

CTMP S UB 2

-3000

RI detector response

6000

-4000

Elution time (min)

CTMP S UB 1

8000

-2000 0

RI detector response

RI detector response

CTMP S B 2

3000
2000
1000
0
-1000 0

10

20

30

-2000
-3000

Elution time (min)

132

Elution time (min)

40

50

Figure 28 (cont’d)

7000

5000

6000

4000

5000

4000
3000
2000
1000
0
-1000 0

5

10 15 20 25 30 35 40 45 50

RI detector response

HPP B 3

RI detector response

HPP B 2

1000
0
0

10

20

30

40

50

-2000

-3000

-3000

Elution time (min)

HPP UB 1

Elution time (min)

HPP UB 2

5000

5000

4000

4000

RI detector response

RI detector response

2000

-1000

-2000

3000
2000
1000
0
-1000 0

5

10 15 20 25 30 35 40 45 50

-2000

5000
4000
3000
2000
1000
0
0

10

20

30

40

50

-2000
-3000

2000
1000
0
-1000 0

-3000

Elution time (min)

HPP UB 3

-1000

3000

5

10 15 20 25 30 35 40 45 50

-2000

-3000

RI detector response

3000

Elution time (min)

133

Elution time (min)

70000

CTMP A B

60000

CTMP A UB
50000

RI detectore response

CTMP S B
CTMP S UB

40000

HPP B
30000
HPP UB
20000

10000

0
0

2

4

6

-10000

8

10

12

14

16

Elution volume (ml)

Figure 29 Size exclusion chromatography whole elution profiles from single column system of
hemicellulose from different approaches, (A: aspen; B: bleached; S: spruce; UB: unbleached).

40000

(a)

18000

Fresh

30000

RI detector response

RI detector response

35000

(b)

20000

5 days later

25000
20000
15000
10000

16000

Fresh

14000

5 days later

12000
10000
8000
6000

4000

5000

2000

0

0
5

6

7

8

5

Elution volume (ml)

6
7
Elution volume (ml)

8

Figure 30 Size exclusion chromatography elution profiles of xylans, freshly made solution and
after 5 days storage, (a) xylan from birchwood; (b) xylan from oat spelts.

134

4

(a)

3.5

Stress (ksi)

3
2.5
2
1.5
1
0.5
0
-0.5

0

1

2

3

4

5

6

7

8

9

10

Stain (%)

1.4

(b)

Tensile load (lbf)

1.2
1
0.8

0.6
0.4
0.2
0
-0.2

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

Extension (in)

Figure 31 Schematic diagrams of MOE calculation, (a) stress-strain method; (b) tensile loadextension method.

Table 21 MOE calculation by three different approaches
Computed maximum
slope (lbf/in)
63.47
77.97
91.62

MOE
Load/extension
MOE
Stress/strain
MOE
(Gpa)
slope (lbf/in)
(Gpa)
slope (ksi)
(Gpa)
1.20
63.52
1.20
177.30
1.22
1.30
74.71
1.25
184.91
1.27
1.31
94.88
1.36
199.87
1.38
1.27
1.27
1.29
0.06
0.08
0.08
Note: Numbers in red frame are the average values of the 3 times calculation of MOE above.
Numbers in green frame are the STDEV values of the 3 times calculation of MOE above.
135

REFERENCES

136

REFERENCES

1.

Agneta F. and K. Björn. 2009. Xylan from Bleached Hardwood Pulp- New Opportunities.
Vtt technical research center of Finland TAPPI Engineering, Pulping &Environmental
Conference, Memphis, Tennessee.

2.

Agneta M., Kuckurek M., Pyiatte J. and E. Wright. 1993. Kraft Pulping. A Compilation of
Notes. TAPPI Press. 6–7.

3.

Agoda-Tandjawa G., Durand S., Berot S., Blassel C., Gaillard C., Garnier C. and J.L.
Doublier. 2010. Rheological characterization of microfibrillated cellulose suspensions after
freezing. Carbohydrate Polymers, 80: 677–686.

4.

Ahvazi B., Radiotis T., Bouchard J. and K. Goel. 2007. Chemical pulping of steam exploded
mixed hardwood chips. Journal of Wood Chemistry Technology, 27 (2): 49–63.

5.

Alekhina M., Mikkonen K.S., Alén R., Tenkanen M. and H. Sixta. 2014.
Carboxymethylation of alkali extracted xylan for preparation of bio-based packaging films.
Carbohydrate Polymers, 100: 89-96.

6.

Allen L., Nelson A.I., Steinberg M.P. and J.N. McGill. 1963. Edible carbohydrate food
coatings. II. Evaluation on fresh meat products. Food Technology, 17: 1442-1448.

7.

Anglès, M.N. and A. Dufresne. 2000. Plasticized starch/tunicin whiskers nanocomposites: 1.
Structural analysis. Macromolecules, 33: 8344–8353.

8.

Antoniewski M.N. and S.A. Barringer. 2010. Meat shelf life and extension using
collagen/gelatin coatings: a review. Critical Reviews in Food Science and Nutrition, 50:
644-653.

9.

Arms K. and P. S. Camp. 1995. Biology 4th Ed. Saunders College Publication.

10. Auras R., Harte B. and S. Selke. 2004. An overview of polylactides as packaging materials.
Macromolecular Bioscience, 4: 835-864.
11. Azizi Samir M.A.S., Alloin F. and A. Dufresne. 2005. Review of recent research into
cellulosic whiskers, their properties and their application in nanocomposite field.
Biomaromolecules, 6 (2): 612-626.
12. Baldwin E.A.. 1994. Edible coatings for fresh fruits and vegetables: past, present and future.
In: Krochta J., Baldwin E. and M. Nisperos-Carriedo (Eds.). Edible Coatings and Films to
Improve Food Quality. Technomic Publishing Company Incorporated, Lancaster, PA, 25-65.
13.

Ban W., Chen X., Andrews G. and A.V. Heiningen. 2011. Influence of Hemicellulose
Pre-extraction and Re-adsorption on pulp physical strength II. Beatability and Strength
Study. Cellulose Chemistry and Technology, 45 (9-10): 633-641.
137

14.

Bastioli C., Lombi R., Deltredici G. and I. Guanella. 1991. De-structuring starch for use
in biodegradable plastics articles-By heating non-dried, non-water- added starch with
plasticizer and enzyme in an extruder. Eur. Patent. EP400531-A1.

15.

Ben H., Mu W., Deng Y. and A.J. Ragauskas. 2013. Production of renewable gasoline
from aqueous phase hydrogenation of lignin pyrolysis oil. Fuel, 103: 1148–1153.

16.

Bhat N.V. and D.N. Makwana. 1988. Effects of Swelling Treatments on fine structure
and mechanical properties of cellophane film. Textile Research Journal, 58: 233-238.

17.

Bickford, J. H. 1995. An introduction to the design and behavior of bolted joints (3rd Ed.)
New York, NY: Taylor & Francis Group.

18.

Binder J.B., Gray M.J., White J.F. Zhang Z.C. and J.E. Holladay. 2009. Reactions of
lignin model compounds inionic liquids. Biomass and Bioenergy, 33: 1122–1130.

19.

Boeriu C.G., Bravo D., Gosselink R.J.A. and J.E.G.Van Dam. 2004. Characterization of
structure-dependent functional properties of lignin with infrared spectroscopy. Industrial
Crops and Products, 20: 205-218.

20.

Boerjan W., Ralph J. and M. Baucher. 2003. Lignin biosynthesis. Annual Review of
Plant Biology, 54 (1): 519–549.

21.

Buranov A.U. and G. Mazza. 2010. Extraction and characterization of hemicelluloses
from flax shives by different methods. Carbohydrate Polymer, 79: 17-25.

22.

Byrom D.. 1994. Polyhydroxyalkanoates. In: Mobley D.P. (Ed.). Plastic from Microbes:
Microbial Synthesis of Polymers and Polymer Precursors. Hanser, Munich, 5-33.

23.

Carlisle R.. 2004. Scientific American Inventions and Discoveries. John Wiley & Sons,
Inc., New Jersey. 338.

24.

Carvalho A.J.F.. 2013. Starch: Major Sources, Properties and Applications as
Thermoplastic Materials. In: Ebnesajjad S. (Ed.). Handbook of Biopolymers and
Biodegradable Plastics: Properties, Processing, and Applications. Waltham, MA, 129-152.

25.

Chanzy H., Dubé M., Marchessault R.H. and J.F. Revol. 1979. Singel-Crystals and
Oriented Crystallization of Ivory Nut Mannan. Biopolymers, 18: 887-898.

26.

Chen L.; Tang C.Y.; Ning N.Y.; Wang C.Y.; Fu Q. and Q. Zhang. 2009. Reparation and
properties of chitosan/lignin composite films. Chinese Journal of Polymer Science, 27 (5):
739-746.

27.

Chinnan M.S. and H.J. Park. 1995. Effect of plasticizer level and temperature on water
vapor transmission of cellulose-based edible films. Journal of Food Process Engineering,
18:417-429.

28.

Cyras V. P., Commisso M. S. and A. Vázquez. 2009. Biocomposites based on renewable
138

resource: acetylated and non-acetylated cellulose
polyhydroxybutyrate. Polymer, 50(26): 6274-6280.

cardboard

coated

with

29.

Debeaufort F., Quezada-Gallo J.A. and A. Voilley. 1998. Edible films and coatings:
Tomorrow's packagings: A review. Critical Reviews in Food Science, 38(4):299–313.

30.

Deniaud E., Fleurence J. and M. Lahaye. 2003. Interactions of the mix-linked β-(1→3)/β(1→4)-D-xylans in the cell wall of palmaria palmata (Rhodophyta). Journal of
Physiology. 39: 74-82.

31.

Dervilly-Pinel G., Tran V. and L. Saulnier. 2004. Investigation of the distribution of
arabinose residues on the xylan backbone of water-soluble arabinoxylans from wheat
flour. Carbohydrate polymers, 55: 171-177.

32.

Dimitris S. Argyropoulos. 2007. Materials, Chemicals, and Energy from Forest Biomass.
ACS Symposium Series; American Chemical Society: Washington, DC.

33.

Dodson B.. 2012. Wood pulp extract stronger than carbon fiber or Kevlar.
http://www.gizmag.com/cellulose-nanocrystals-stronger-carbon-fiber-kevlar/23959/.

34.

Domenek S., Courgneau C. and V. Ducruet. 2011. Characteristics and Applications of
PLA. In: S. Kalia, L. Avérous (Eds.). Biopolymers: Biomedical and Environmental
Applications, first ed. Wiley-Scrivener, New-York, 183-224.

35.

Doroudgarian N.. 2011. Nanocomposite Films of Hemicellulose and Nanocelluloses with
Improved Barrier Properties for Packaging Applications. Department of Engineering
Sciences and Mathematics, LuleåUniversity of Technology.

36.

Dumitriu S. 1996. In Polysaccharides in medicinal applications. CRC Press LLC, Boca
Raton, FL, USA.

37.

Ebnesajjad S. (Ed.). 2013. Handbook of Biopolymers and Biodegradable Plastics:
Properties, Processing, and Applications. Waltham, MA, 109-128.

38.

Ebringerov A., Hrom dkov Z., Kačur kov M. and M. Antal. 1994. Quaternized
xylans: synthesis and structural characterization. Carbohydrate Polymers, 24 (4): 301–
308.

39.

Ebringerová A. and T. Heinze. 2000. Xylan and xylan derivatives – biopolymers with
valuable properties, 1. Naturally occurring xylans structures, isolation procedures and
properties. Macromolecular Rapid Communications, 21(9):542-556.

40.

Ebringerová A., Hromadkova, Z. and T. Heinze. 2005. Polysaccharides 1: structure,
characterization and use. In: Heinze T. (Eds.). Hemicellulose. Springer-verlag Berlin,
Heidelberger Platz 3, D-14197 Berilin, Germany. 1-67.

41.

Edlund U., Zhu Y. R., and A.C. Albertsson. 2010. Barrier Films from Renewable
Forestry Waste. Biomacromolecules, 11:2532–2538.
139

42.

Egüés I., Eceiza A. and J. Labidi. 2013. Effect of different hemicelluloses characteristics
on film forming properties. Industrial Crops and Products, 47: 331-338.

43.

Environmental Protection Agency (EPA). 2007. Reregistration eligibility decision for
glutaraldehyde. EPA 739-R-07-006. Washington D.C. US.

44.

Fang Q. and M.A. Hanna. 2000. Functional properties of polylactic acid starch-based
loose-fill packaging foams. Cereal Chemistry, 77: 779-783.

45.

Fang Q. and M.A. Hanna. 2001. Preparation and characterization of biodegradable
copolyester-starch based foams. Bioresource Technology, 78: 115-122.

46.

FAO (Food and Agriculture Organization of the United Nations). Pulp, paper and
paperboard capacity survey. 2012-2017, Rome, Italy.

47.

FAO (Food and Agriculture Organization of the United Nations). 2007. FAOSTAT
Statistical database Agriculture, Rome, Italy.

48.

FAO (Food and Agriculture Organization of the United Nations). 2010. Global Forest
Resources Assessment, Rome, Italy.

49.

Félix M., Martín-Alfonso J.E., Romero A. and A. Guerrero. 2014. Development of
albumen/soy biobased plastic materials processed by injection molding. Journal of Food
Engineering, 125: 7–16.

50.

Felton L.A. and S.C. Porter. 2013. An update on pharmaceutical film coating for drug
delivery. Expert opinion on drug delivery, 10(4): 421-435.

51.

Fernandez-Bolanos J., Felizon B., Heredia A., Guillen R., and A. Jimenez. 1999.
Characterization of the lignin obtained by alkaline delignification and of the cellulose
residue from steam-exploded olive stones. Bioresource technology, 68: 121-132.

52.

Fiebach K. and D. Grimm. 2000. Resins, Natural. Ullmann's Encyclopedia of Industrial
Chemistry.

53.

Food and Drug Administration (FDA). 2014. CFR-Code of Federal Regulations Title 21.
21CFR176.170. Silver Spring, MD, USA.

54.

Freudenberg K. and A.C. Neish. 1968. Constitution and Biosynthesis of Lignin. SpringVerlag. Berlin.

55.

Gabrieliia I., Gatenholma P., Glasserb W.G., Jainb R.K and L. Kennec. 2000. Separation,
characterization and hydrogel-formation of hemicellulose from aspen wood.
Carbohydrate Polymers, 43: 367–374.

56.

Gallant D.J., Bouchet B. and P.M. Baldwin. 1997. Microscopy of starch: evidence of a
new level of granule organization. Carbohydrate Polymer, 32: 177-191.

140

57.

Galliard T. and P. Bowler. 1987. Morphology and composition of starch. In: Galliard T.
(Ed.). Starch: Properties and Potential, 1st Ed.. John Wiley & Sons, New York. 55-78.

58.

Gao W.H., Chen K.F., Xiang Z.Y., Yang F., Zeng P., Li J., Yang R.D., Rao G.H. and H.
Tao. 2013. Kinetic study on pyrolysis of tobacco residues from the cigarette industry.
Industrial Crops and Products, 44: 152-157.

59.

García, M. A.; Pinotti, A.; Martino, M. N.; Zaritzky, N. E. 2009. Characterization of
Starch and Composite Edible Films and Coatings. In Huberand, K. C., Embuscado, M. E.,
(Eds), Edible Films and Coatings for Food Applications; Springer: New York: 169–209.

60.

Garlotta D.. 2001. A literature review of poly (lactic acid). Journal of Polymers and the
Environment, 9: 63-84.

61.

Gatenholm, P. and M. Tenkanen. 2004. Hemicelluloses: Science and Technology. Oxford
University Press, New York.

62.

Georget D. M., Ng A., Smith A. C. and K. W. Waldron. 1999. Thermal characterization
of oxidatively cross-linked American corn bran hemicellulose. Journal of the Science of
Food and Agriculture, 79:481-483.

63.

Gosselink, R.J.A.; De Jong, E.; Guran, B. and A. Abächerli. 2004. Co-ordination network
for lignin—standardisation, production and applications adapted to market requirements
(EUROLIGNIN). Industrial Crops and Products, 20: 121-129.

64.

Greener-Donhowe I. and O. Fennema. 1993. The Effects of Plasticizers on crystallinity,
Permeability, and Mechanical Properties of Methylcellulose films. Journal of Food
Processing and Preservation, 17: 231-246.

65.

Gröndahl M., Eriksson L. and P. Gatenholm. 2004. Material Properties of Plasticized
Hardwood Xylans for Potential Application as Oxygen Barrier Films.
Biomacromolecules, 5: 1528-1535

66.

Grubb H.. 2010. Xylophane- A sustainable barrier for packaging. TAPPI, New Mexico,
USA.

67.

Gruber P.R., Hall E.S., Kolstad J.H., Iwen M.L., Benson R.D. and R.L. Borchardt. 1992.
US Patent No. 5142023.

68.

Habibi Y., Lucia L.A., and O.J. Rojas. 2010. Cellulose nanocrystals: chemistry, selfassembly and applications. Chemical reviews, 110: 3479-3500.

69.

Hagenmaier R.D. and P.E. Shaw. 1990. Moisture permeability of edible films made with
fatty acid (hydroxypropyl) methylcellulose. Journal of Agricultural and Food Chemistry,
38: 1799-1803.

70.

Hair J.F., Black W.C., Babin B.J. and R.E. Anderson. 2009. Multivariate Data Analysis
(7th Ed.). Prentice Hall. Upper Saddle River, NJ, US.
141

71.

Hansen N.M.L., Blomfeldt T.O.J., Hedenqvist M.S. and D.V. Plackett. 2012. Properties
of plasticized composite films prepared from nanofibrillated cellulose and birch wood
xylan. Cellulose, 19: 2015–2031.

72.

Hansen N.M.L. and D. Plackett. 2008. Sustainable Films and Coatings from
Hemicelluloses: A Review. Biomacromolecules, 9(6):1493-1505.

73.

Hartman J.. 2006. Hemicellulose as Barrier Material. KTH, Department of Fiber and
Polymer Technology, School of Chemical Science and Engineering, Royal Institute of
Technology. Stockholm, Sweden.

74.

Hartman J., Albertsson A.C., and J. Sjöberg. 2006. Surface- and Bulk-Modified
Galactoglucomannan Hemicellulose Films and Film Laminates for Versatile Oxygen
Barriers. Biomacromolecules, 7: 1983-1989

75.

Heikkinen S.L., Mikkonen K.S., Pirkkalainen K., Serimaa R., Joly C. and M. Tenkanen.
2013. Specific enzymatic tailoring of wheat arabinoxylan reveals the role of substitution
on xylan film properties. Carbohydrate polymers, 92: 733-740.

76.

Heikkinen S. L., Mikkonen K. S., Koivisto P., HeikkiläM. I., Pirkkalainen K., Liljeström
V., Serimaa R. and M. Tenkanen. 2014. Long-Term Physical Stability of Plasticized
Hemicellulose Films. BioResources, 9: 906-921.

77.

Higson, A. and C. Smith. 2011. NNFCC Renewable Chemicals Factsheet: Lignin. The
bioeconomy consultants.

78.

Hirvikorpi T., Vähä-Nissi M., Nikkola J., Harlin A. and M. Karppinen. 2011. Thin Al2O3
barrier coatings onto temperature-sensitive packaging materials by atomic layer
deposition. Surface & Coatings Technology, 205: 5088–5092

79.

Höije A., Sternemalm E., Heikkinen S., Tenkanen M., and P. Gatenholm. 2008. Material
Properties of Films from Enzymatically Tailored Arabinoxylans. Biomacromolecules, 9:
2042–2047.

80.

Holladay J.E., Bozell J.J., White J.F. and D. Johnson. 2007. Top value-added chemicals
from biomass. Volume II—results of screening for potential candidates from biorefinery
lignin. PNNL-16983. Richland, WA: Pacific Northwest National Laboratory.

81.

Huneault M.A. and H. Li. 2007. Morphology and properties of compatibilized
polylactide/thermoplastic starch blends. Polymer, 48: 270-280.

82.

Hüttermann A., Mai C. and A. Kharazipour. 2001. Modification of lignin for the
production of new compounded materials. Applied Microbiology and Biotechnology, 55:
387–394.

83.

Isogai A., Saitoa T. and
Nanoscale, 3: 71-85.

H. Fukuzumia. 2011. TEMPO-oxidized cellulose nanofibers.

142

84.

Jackson M. G. 1977. The alkali treatment of straws. Animal Feed Science and
Technology, 2: 105–130.

85.

Jacobs A. and O. Dahlman. 2001. Characterization of the molar masses of hemicelluloses
from wood and pulps employing size exclusion chromatography and matrix-assisted laser
desorption ionization time-of-flight mass spectrometry. Biomacromolecules, 2: 894-905.

86.

Jain R.K., Sjӧstedt M. and W.G. Glasser. 2001. Thermoplastic xylan derivatives with
propylene oxide. Cellulose, 7: 319–336.

87.

Jayakrishnan A. and S.R. Jameela. 1996. Glutaraldehyde as a fixative in bioprostheses
and drug delivery matrices. Biomaterials, 471-484.

88.

Jenkins P.J., Cameron R.E. and A.M. Donald. 1993. A universal feature in the structure
of starch granules from different botanical sources. Starch/Stärke, 45: 417-420.

89.

Jiang G., Nowakowski D.J. and A.V. Bridwater. 2010. A systematic study of the kinetics
of lignin pyrolysis. Thermochimica Acta, 498 (1-2): 61–66 (a).

90.

Jiang G., Nowakowski D.J. and A.V. Bridgwater. 2010. Effect of the temperature on the
composition of lignin pyrolysis products. Energy Fuels, 24: 4470–4475 (b).

91.

Jiang L. and J. Zhang. 2013. Biodegradable polymers and polymer blends. In: Ebnesajjad
S. (Ed.). Handbook of Biopolymers and Biodegradable Plastics: Properties, Processing,
and Applications. Waltham, MA, 109-128.

92.

Jin X.J. and D.P. Kamdem. 2009. Chemical composition, crystallinity and crystallite
cellulose size in populous hybrids and aspen. Cellulose Chemistry Technology, 43 (7-8):
229-234.

93.

Kainuma K.. 1984. Starch oligosaccharides: linear, branched, and cyclic. In: Whistler
R.L., Bemiller J.N. and E.F. Paschall (Eds.). Starch: Chemistry and Technology, 2nd Ed..
Academic Press, San Diego, CA. 125-152.

94.

Kang H.J., Jo C., Kwon J.H., Kim J.H., Chung H.J. and M.W. Byun. 2007. Effect of a
pectin based edible coating containing green tea powder on the quality of irradiated pork
patty. Food Control, 18: 430-435.

95.

Kang S.M., Li X.L., Fan J. and J. Chang. 2013. Hydrothermal conversion of lignin: A
review. Renewable and Sustainable Energy Reviews, 27: 546-558.

96.

Kansal S.K., Singh M. and D. Sud. 2008. StudiesonTiO2/ZnO photo catalyzed
degradation of lignin. Journal of Hazardous Materials, 153: 412–417.

97.

Katherine E.. 1977. Anatomy of Seed Plants. John Wiley and Sons (WIE), New York.

98.

Kennedy J.F., Cabral J.M.S., Sa´-Correia I. and C.A. White. 1987. Starch biomass: a
chemical feedstock for enzyme and fermentation processes. In: Galliard T. (Ed.). Starch:
143

Properties and Potential, 1st Ed.. John Wiley & Sons, New York. 115-148.
99.

Keshri C.R.C. and M.K. Sanyal. 2009. Effect of sodium alginate coating with
preservatives on the quality of meat patties during refrigerated (4±1℃) storage. Journal
of Muscle Foods, 20: 275-292.

100.

Kirwan M.J.. 2003. Paper and paperboard packaging. In: Coles R., Mcdowell D. and M.J.
Kirwan (Eds.). Food Packaging Technology. Blackwell Publishing Ltd, London. 241-281.

101.

Kirwan M.J. and J.W. Strawbridge. 2003. Plastics in food packaging. In: Coles R.,
Mcdowell D. and M.J. Kirwan (Eds.). Food Packaging Technology. Blackwell Publishing
Ltd, London. 174-240.

102.

Kochumalayil J.J., Zhou Q., Kasai W. and L.A. Berglund. 2013. Regioselective
modification of a xyloglucan hemicellulose for high-performance biopolymer barrier
films. Carbohydrate Polymers, 93: 466– 472.

103.

Koran Z. and D.P. Kamdem. 1989. The bending stiffness of paperboard. TAPPI journal.
72: 175-179.

104.

Koutinas A.A., Wang R. and C. Webb. 2004. Evaluation of wheat as generic feedstock
for chemical production. Industrial Crops and Products, 20: 75-88.

105.

Krawczyk H., Persson T., Andersson A., and A.S. Jönsson. 2008. Isolation of
hemicelluloses from barley husks. Food and Bioproducts Processing, 86: 31-36

106.

Krawczyk H., and A.S. Jönsson. 2011. Separation of dispersed substances and
galactoglucomannan in thermomechanical pulp process water by microfiltration.
Separation and Purification Technology, 79: 43–49.

107.

Kuntz E.. 1964. Preparation of collagenous materials. US Patent No. 3152976.

108.

Kuorwel K.K., Cran M.J., Sonneveld K., Miltz J. and S.W. Bigger. 2013. Water Sorption
and Physicomechanical Properties of Corn Starch-Based Films. Journal of Applied
Polymer Science, 128(1):530-536.

109.

Laine C., Harlin A., Hartman J., Hyvärinen S., Kammiovirta K., Krogerus B., Pajari H.,
Rautkoski H., Setälä H., Sievänen J., Uotila J. and M. Vähä-Nissi. 2013.
Hydroxyalkylated xylans – Their synthesis and application in coatings for packaging and
paper. Industrial Crops and Products, 44: 692– 704.

110.

Lai V. M. F., Lu S., He W. H. and H. H. Chen. 2006. Non-starch polysaccharide
compositions of rice grains with respect to rice variety and degree of milling. Food
Chemistry, 101: 1205–1210.

111.

Li W. L., Coffin D. R., Jin T. Z., Latona N., Liu C.K., Liu B., Zhang J.W. and L.S. Liu.
2012. Biodegradable Composites from Polyester and Sugar Beet Pulp with Antimicrobial
Coating for Food Packaging. Journal of Applied Polymer Science, 126: E361–E372.
144

112.

Lindblad, M. S. and A. C. Albertsson. 2005. In Polysaccharides: Structural diVersity and
functional Versatility; Dumitriu, S., Ed.; Marcel Dekker: New York, 491–508.

113.

Lindström and István Siro. 2010. Physical Properties and Morphology of Films Prepared
from Microfibrillated Cellulose and Microfibrillated Cellulose in Combination with
Amylopectin. Journal of Applied Polymer Science, 117: 3601–3609

114.

Liu W., Hou Q., Mao C., Yuan Z., and K. Li. 2012. Hemicelluloses Prior to Aspen
Chemithermomechanical Pulping: Pre- Extraction, Separation, and Characterization.
Journal of Agricultural and Food Chemistry, 60: 4880−4885.

115.

Lundqvist J., Teleman A., Junel L., Zacchi G., Dahlman O., Tjerneld F. and H. Stalbrand.
2002. Carbohydrate Polymer, 48: 29-39.

116.

Luo X.G. 2008. The extraction of high quality lignin and its thermoplastic modification.
Beijing. Chemical industry press.

117.

Mabesa R.C., Marshall R.T. and M.E. Anderson. 1980. Milk films exposed to high
humidity: studies with electron microscopy and electrophoresis. Journal of Food
Protection, 43: 29-35.

118.

Malawer E.G. and L. Senak. 2003. Introductionto size exclusion chromatography. In: Wu
C.S. (Eds). Handbook of Size Exclutsion Chromatography and Related Techniques.
Marcel Dekker Inc., NY, 1-24.

119.

Martín-Closas L. and A.M Pelacho. 2011. Agronomic potential of biopolymer films. In:
Plackett, D. (Eds). Biopolymers: New Materials for Sustainable Films and Coatings. John
Wiley & Sons, Hoboken, NJ, 277-299.

120.

Martín-Sampedro R., Eugenio M.E., Revilla E., Martín J.A. and J.C. Villar. 2011a.
Integration of kraft pulping on a biorefinery by the addition of a steam explosion
pretreatment. Bioresources, 6 (1): 513–528.

121.

Martín-Sampedro R., Martin J.A., Eugenio M.E., Revilla E. and J.C. Villar. 2011b.
Steam explosion treatment of Eucalyptus globulus wood: influence of operational
conditions on chemical and structural modifications. Bioresources, 6 (4): 4922–4935.

122.

Martín-Sampedro R., Eugenio M.E., Moreno J.A., Revilla E. and J.C. Villar. 2014.
Integration of a kraft pulping mill into a forest biorefinery: Pre-extraction of
hemicellulose by steam explosion versus steam treatment. Bioresource Technology, 153:
236–244.

123.

Matveev Y., Grinberg V. and V. Tolstoguzov. 2000. The plasticizing effect of water on
proteins, polysaccharides and their mixtures. Glassy state of biopolymers, food and seeds.
Food Hydrocolloid, 14: 425–437.

145

124.

McHugh T.H., Avena-Bustillos R. and J.M. Krochta. 1993. Hydrophilic edible films:
modified procedure for water vapor permeability and explanation of thickness effects.
Journal of Food Science, 58: 899-903.

125.

Mesquita J., Donnici C.L. and F. V. Pereira. 2010. Biobased Nanocomposites from
Layer-by-Layer Assembly of Cellulose Nanowhiskers with Chitosan. Biomacromolecules,
11: 473–480

126.

Mikkelsen D., Gidley M.J. and B.A. Williams. 2011. In Vitro Fermentation of Bacterial
Cellulose Composites as Model Dietary Fibers. Journal of Agricultural and Food
Chemistry, 59: 4025-4032.

127.

Mikkonen K.S., Heikkinen S., Soovre A., Peura M., Serimaa R., Klemens F.R.A. and G.
Dieter. 2000. Resins, Natural. Ullmann's Encyclopedia of Industrial Chemistry.

128.

Mikkonen K. S., Pitkänen L., Liljeström V., Bergström E. M., Serimaa R., Salmén L. and
M. Tenkanen. 2012. Arabinoxylan structure affects the reinforcement of films by
microfibrillated cellulose. Cellulose, 19: 467-480.

129.

Mikkonen K.S. and M. Tenkanen. 2012. Sustainable foodpackaging materials based on
future biorefinery products: Xylans and mannans (Review). Trends in Food Science &
Technology, 28: 90-102.

130.

Mikkonen K.S., Heikkinen S., Soovre A., Peura M., Serimaa R., Talja R.A., Helén H.,
Hyvönen L. and M. Tenkanen. 2009. Films from Oat Spelt Arabinoxylan Plasticized with
Glycerol and Sorbitol. Journal of Applied Polymer Science, 114: 457–466

131.

Milda E. Embuscado and Kerry C. Huber. 2009. Edible Films and Coatings for Food
Applications. Springer Dordrecht Heidelberg London, New York

132.

Morohoshi N.. 1991. Chemical Characterization of Wood and its Components. In: Hon
D.N.S. and N. Shiraishi. (Ed). Wood and cellulosic chemistry. Marcel Dekker Inc., New
York, 331-394.

133.

Mu W., Ben H., Ragauskas A. and Y. Deng. 2013. Lignin Pyrolysis Components and
Upgrading-Technology Review. Bioenergy research, 4 (6): 1183-1204.

134.

Nussinovitch A.. 2013. Biopolymer films and composite coatings. In: Ebnesajjad S. (Ed.).
Handbook of Biopolymers and Biodegradable Plastics: Properties, Processing, and
Applications. Waltham, MA, 295-327.

135.

Oksanen A., Rantanen T., Retulainen E., Salminen K. and H. Brumer. 2011. Improving
wet web runnability and paper quality by an uncharged polysaccharide. Journal of
Biobased Materials and Bioenergy, 5: 187-191.

136.

Oliveira L., Evtuguin D. V., Cordeiro N., Silvestre A. J., Silva A. M. and I. C. Torres.
2006. Structural characterization of lignin from leaf sheaths of ―Dwarf cavendish‖ banana
plant. Journal of agricultural and food chemistry, 54: 2598-2605.
146

137.

Otey F.H. and W.M. Doane. 1984. Chemicals from starch. In: Whistler R.L., Bemiller
J.N. and E.F. Paschall (Eds.). Starch: Chemistry and Technology, 2nd Ed.. Academic
Press, San Diego, CA. 389-416.

138.

Ouyang X., Lin Z., Deng Y., Yang D. and X. Qiu. 2010. Oxidative degradation of soda
lignin assisted by microwave irradiation. Chinese Journal of Chemical Engineering, 18:
695–702.

139.

Pääkkö M., Vapaavuori J., Silvernnoinen R., Kosonen H., Ankerfors M., Lindstrom T.,
Berglund L.A. and O. Ikkala. 2008. Long and entangled native cellulose I nanofibers
allow flexible aerogels and hierarchically porous templates for functionalities. Soft
Matter, 4(12): 2492-2499.

140.

Palm, M. and G. Zacchi. 2003. Extraction of hemicellulosic oligosaccharides from spruce
using microwave oven or steam treatment. Biomacromolecules, 4: 617-623.

141.

Parulekar Y. and A.K. Mohanty. 2007. Extruded biodegradable cast films from
polyhydroxyalkanoate and thermoplastic starch blends: fabrication and characterization.
Macromolecular Materials and Engineering, 292: 1218-1228.

142.

Peelmana N., Ragaert P., Meulenaer B.D., Adons D., Peeters R., Cardon L., Impe F.V.
and F. Devlieghere. 2013. Application of bioplastics for food packaging. Trends in Food
Science and Technology, 32: 128-141.

143.

Peng F., Ren J., Xu F., Bian J., Peng P., and R. Sun. 2009. Comparative Study of
Hemicelluloses Obtained by Graded Ethanol Precipitation from Sugarcane Bagasse.
Journal of Agricultural and Food Chemistry, 57: 6305–6317.

144.

Peng X., Bian J., Li M., Xiao X., Xia X-L., Yin W., and R. Sun. 2013. Graded Ethanol
Fractionation and Stuctural Characterization of Alkali-Extractable Hemicellloses from
Olea europaea L.. BioResources, 8(1): 1110-1123.

145.

Peng X., Bian J., Peng P., Guan Y., Xu F. and R. Sun. 2013. Fractional separation and
structural features of hemicelluloses from sweet sorghum leaves. Bioresources, 7: 47444759.

146.

Péroval C., Debeaufort F., Despre D. and A. Voilley. 2002. Edible arabinoxylan-based
films. 1. Effect of lipid type on water vapor permeability, film structure, and other
physical characteristics. Journal of Agricultural and Food Chemistry, 50: 3977-3983.

147.

Persson T., Jönsson, A.S. and G. Zacchi. 2005. In 14th European Biomass Conference
and Exhibition.

148.

Persson T., Nordin A. K., Zacchi G. and A.S. JÖnsson. 2007. Economic Evaluation of
Isolation of Hemicelluloses from Process Streams from Thermomechanical Pulping of
Spruce. Applied Biochemistry and Biotechnology, 136–140: 741-752.

147

149.

Petzold-Welcke K., Schwikal K., Daus S. and T. Heinze. 2014. Xylan derivatives and
their application potential – Mini-review of own results. Carbohydrate Polymers. 100:
80-88.

150.

Peura M., Karppinen T., Soovre A., Salmi A., Tenkanen Ma., Hæggström E. and R.
Serimaa. 2008. Crystallization and shear modulus of a forming biopolymer film
determined by in situ x-ray diffraction and ultrasound reflection methods. Journal of
applied physics, 104: 023513

151.

Phan The D., Debeaufort F., Peroval C., Courthaudon J.L. and A. Voilley. 2002.
Arabinoxylan lipid-based edible films and coatings: 3. Influence of drying temperature on
film structure and functional properties. Journal of Agricultural and Food Chemistry. 50:
2423-2428.

152.

Phan The D., Debeaufort F., Voilley A. and D. Luu. 2009. Biopolymer interactions affect
the functional properties of edible films based on agar, cassava starch and arabinoxylan
blends. Journal of Food Engineering, 90: 548-558.

153.

Plackett D., Anturi H., Hedenqvist M., Ankerfors M., Gällstedt M., EbringerováA. M.,
HromádkováZ.and T. Heinze. 2005. Hemicellulose. Adv. Polym. Sci. 186: 1–67

154.

Plackett D., Anturi H., Hedenqvist M., Ankerfors M., Gällstedt M., Lindström T. and I.
Siro. 2010. Physical Properties and Morphology of Films Prepared from Microfibrillated
Cellulose and Microfibrillated Cellulose in Combination with Amylopectin. Journal of
Applied Polymer Science. 117: 3601–3609.

155.

Plackett D. (Eds). 2011. Biopolymers: New Materials for Sustainable Films and Coatings.
John Wiley &Sons, Hoboken, NJ.

156.

Robertson G.H., Wong D.W.S., Lee C.C., Wagschal K.W., Smith M.R. and W.J. Orts.
2006. Native or raw starch digestion: a key step in energy efficient biorefining of grain.
Journal of Agricultural and Food Chemistry. 54: 353-365.

157.

Robertson, G. 2008. State-of-the-art biobased food packaging meterials. In: Chiellini E.
(Eds.). Environmentally Compatible Food Packaging. Woodhead Publishing
limited:Cambridge, UK:3–28.

158.

Robinson C.. 1953. The hot and cold forms of gelatin. In: Randel J.T. (Ed.). Nature and
Structure of Collagen. Academic Press, New York. 96-105.

159.

Robinson J.W., Frame E.M.S. and F.M. Frame II. 2005. Visible and Ultraviolet Moleclar
Spectroscopy. In: Robinson J.W., Frame E.M.S. and F.M. Frame II (Ed.). Undergraduate
instrumental analysis, 6th edition. Marcel Dekker Inc., New York, 317-379.

160.

Rodríguez-González F.J., Ramsay B.A. and B.D. Favis. 2003. High performance
LDPE/thermoplastic starch blends: a sustainable alternative to pure polyethylene.
Polymer, 44: 1517-1526.

148

161.

Rodr guez-Gutiérrez G., Rubio-Senent F., Lama-Mu oz A., Garc a A., and J. Fern ndezBola os. 2014. Properties of Lignin, Cellulose, and Hemicelluloses Isolated from Olive
Cake and Olive Stones: Binding of Water, Oil, Bile Acids, and Glucose. Journal of
agricultural and food chemistry, 62: 8973-8981.

162.

Roubroeks J. P., Andersson R. and P. Åman. 2000. Structural features of (1→3),(1→4)β-D-glucan and arabinoxylan fractions isolated from rye bran. Carbohydrate Polymer,
42:3-11.

163.

Rowell R. M.. 2012. Handbook of wood chemistry and wood composites, 2nd edition.
CRC Press LLC, Boca Raton, FL, USA

164.

Rudnik E.. 2013. (a) Compostable Polymer Materials: Definitions, Structures, and
Methods of Preparation. In: Ebnesajjad S. (Ed.). Handbook of Biopolymers and
Biodegradable Plastics: Properties, Processing, and Applications. Waltham, MA, 189-211.

165.

Rudnik E.. 2013. (b) Biodegradability Testing of Compostable Polymer Materials. In:
Ebnesajjad S. (Ed.). Handbook of Biopolymers and Biodegradable Plastics: Properties,
Processing, and Applications. Waltham, MA, 213-263.

166.

Ruzene D.S., Silva D.P., Vicente A.A., Gonçalves A.R. and J.A. Teixeira. 2008. An
Alternative Application to the Portuguese Agro-Industrial Residue: Wheat Straw. Applied
Biochemistry Biotechnology. 147: 85–96.

167.

Saadatmand S., Edlund U., Albertsson A.C., Danielsson S., and O. Dahlman. 2012.
Prehydrolysis in Softwood Pulping Produces a Valuable Biorefinery Fraction for Material
Utilization. Environmental Science and Technology, 46: 8389−8396

168.

Sannigrahi P., Ragauskas A.J. and G.A. Tuskan. 2010. Poplar as a feedstock for biofuels:
A review of compositional characteristics. Biofuels, Bioproducts, Biorefining, 4: 209–
226.

169.

Sargent J.A.. 2007. Dialysis in the 1960s and the first hollow fiber dialyzer. International
journal of artificial organs. 30(11): 953-963.

170.

S rossy z., Blomfeldt T.O.J., Hedenqvist M. S., Koch C.B., Ray S. S., and D. Plackett.
2012. Composite Films of Arabinoxylan and Fibrous Sepiolite: Morphological,
Mechanical, and Barrier Properties. ACS Applied Materials & Interfaces, 4: 3378−3386

171.

Sato S., Gondo D., Wada T., Kanehashi S. and K. Nagai. 2013. Effects of various liquid
organic solvents on solvent-induced crystallization of amorphous poly (lactic acid) film.
Journal of Applied Polymer Science, 129: 1607-1617.

172.

Scheller H.V. and P. Ulvskov. 2010. Hemicelluloses. Annual Review of Plant Biology, 61:
263-289.

149

173.

Schönberg C., Oksanen T., Suurnäkki A., Kettunen H. and J. Buchert. 2001. The
Importance of Xylan for the Strength Properties of Spruce Kraft Pulp Fibers.
Holzforschung, 55: 639-644.

174.

Schröder R., Nicolas P., Vincent S.J.F., Fischer M., Reymond S. and R.J. Redgwell. 2001.
Purification and characterization of a galactoglucomannan from kiwifruit (Actinidia
deliciosa). Carbohydrate Research, 331: 291-306.

175.

Schulze E.. 1891. Zur Kenntnis der chemischen Zusammensetzung der pflanzlichen
Zellmembranen. Berichte der Deutschen Chemischen Gessllschaft, 24: 2277-2287.

176.

Seca A.M.L., Cavaleiro J.A.S., Domingues F.M.J., Silvestre A.J.D., Evtuguin D. and C.P.
Neto. 1998. Structural Characterization of the Bark and Core Lignins from Kenaf
(Hibiscus cannabinus). Journal of Agriculture and food chemistry, 46: 3100-3108.

177.

Selke S., Culter J. and R. Hernandez. 2004. Plastics Packaging: Properties, Processing,
Applications and Regulations, 2nd ed. Hanser Publications, Cincinnati, OH, USA.

178.

Sensidoni A. and D. Peressini. 1997. Edible films: potential innovation for fish products.
Industrie Alimentari, 36: 129-133.

179.

Settineri.W.J. and R.H. Marchessault. 1965. Derivation of possible chain conformations
for poly-β-1,4-anhydroxylose. Journal of Polymer Science, Part C: Polumer Symposia,
11: 253-264.

180.

Shen L., Haufe J. and M. K. Patel. 2009. Product overview and market projection of
emerging bio-based plastics. European polysaccharide network of excellence and
European bioplastics. Utrecht, the Netherlands.

181.

Shi W. and M.J. Dumont. 2014. Review: bio-based films from zein, keratin, pea, and
rapeseed feedstocks. Journal of materials science, 49:1915–1930.

182.

Shukla. R.. and M. Cheryan. 2001. Zein: The industrial protein from corn. Industrial
Crops and Products, 13: 171-192.

183.

Šimkovic I., Nu ez A., Strahan G.D., Yadav M.P., Mendichi R. and K.B. Hicks. 2009.
Fractionation of sugar beet pulp by introducing ion-exchange groups. Carbohydrate
Polymers, 78: 806–812.

184.

Singh A., Yadav K. and A.K. Sen. 2012. Sal (Shorea Robusta) Leaves Lignin Epoxidation
and Its Use in Epoxy Based Coatings. American Journal of Polymer Science, 2: 14-18.

185.

Sjöström E..1993. Wood Chemistry: Fundamentals and Applications, 2nd ed. Academic
press: San Diego, CA, USA.

186.

Smith W.B., Miles P.D., Perry C.H. and S.A. Pugh. 2009. Forest Resources of the United
States, 2007. Gen. Tech. Rep. WO-78. Washington, DC: U.S. Department of Agriculture,
Forest Service, Washington Office. 336 p.
150

187.

Sohn S., Chang S. and I. Hwang. 2003. The effects of NaOH and corona treatments on
triacetyl cellulose and liquid crystal films used in LCD devices. Journal of Adhesion
Science and Technology, 17(3): 453-469.

188.

Sommer I. and Kunz P. M.. 2012. Improving the Water Resistance of Biodegradable
Collagen Films. Journal of Applied Polymer Science, 125: E27–E41.

189.

Soradech S., Nunthanid J., Limmatvapirat S. and M. Luangtana-anan. 2012. An approach
for the enhancement of the mechanical properties and film coating efficiency of shellac
by the formation of composite films based on shellac and gelatin. Journal of Food
Engineering, 108: 94-102.

190.

Spence K. L., Venditti R. A., Rojas O.J., Habibi Y. and J. J. Pawlak. 2010. The effect of
chemical composition on microfibrillar cellulose films from wood pulps: water
interactions and physical properties for packaging applications. Cellulose, 17:835–848.

191.

Spencer R.R. and D.E. Akin. 1980. Rumen microbial-degradation of potassium
hydroxide treated coastal Bermuda grass leaf blades examined by electron-microscopy.
Journal of Animal Science, 51: 1189–1196.

192.

Spiridon I. and V.I. Popa. 2008. Hemicellulose: major sources, properties and
applications. In: Belgacem M.N. and A. Gandini (Eds.). Monomers. Polymers and
Composites from Renewable Resources. 289-304. Elsevier, Amsterdam, the Netherlands.

193.

Stanssens D., denAbbeele V.H., Vonck L., Schoukens G., Deconinck M. and P. Samyn.
2011. Creating water-repellent and super-hydrophobic cellulose substrates by deposition
of organic nanoparticles. Materials Letters, 65: 1781–1784.

194.

Steinbüchel A.. 1991. Polyhydroxyalkanoic acids. In: Byrom D. (Ed.). Biomaterials.
Novel Materials from Biological Sources. Stockton, New York, 124-213.

195.

Stephen A.M.. 1983. In: Aspinall G.O. (Ed). The Polysaccharides. vol. 2. Academic press,
New York, 97-193.

196.

Stepan A.M., Hoije A., Schols H. A., de Waard P. and P. Gatenholm. 2012. Arabinose
Content of Arabinoxylans Contributes to Flexibility of Acetylated Arabinoxylan Films.
Journal of Applied Polymer Science, 125: 2348–2355.

197.

Stevanic J. S., Bergström E. M, Gatenholm P, Berglund L. and L. Salme´n. 2012.
Arabinoxylan/nanofibrillated cellulose composite films. Journal Materials Science, 47:
6724–6732.

198.

Stoklosa R.J. and D.B. Hodge. 2012. Extraction, Recovery, and Characterization of
Hardwood and Grass Hemicelluloses for Integration into Biorefining Processes.
Industrial & Engineering Chemistry Research, 51:11045-11053.

199.

Sun R., Fanga J. M., Tomkinson J. and C.A.S. Hill. 1999. Esterification of
Hemicelluloses from Poplar Chips in Homogenous Solution of N, N151

Dimethylformamide/ Lithium Chloride. Journal of Wood Chemistry and Technology,
19(4): 287–306.
200.

Sun R., Fang J., Tomkinson J., Geng Z. and J. Liu. 2001. Fractional isolation, physicochemical characterization and homogeneous esterification of hemicelluloses from fastgrowing poplar wood. Carbohydrate Polymers, 44: 29–39.

201.

Sun R., Sun X. and J. TomkinsonIn. 2004. Hemicellulose and their derivatives. In:
Gatenholm, P. and M. Tenkanen (Eds.). Hemicellulose: Science and Technology.
American Chemical Society: Washington, DC, 2–22.

202.

Sun R. and S. Hughes. 1998. Fractional extraction and physico-chemical characterization
of hemicelluloses and cellulose from sugar beet pulp. Carbohydrate Polymers, 36: 293–
299.

203.

Sun S., Wen J., Ma M. and R. Sun. 2013. Successive alkali extraction and structural
characterization of hemicelluloses from sweet sorghum stem. Carbohydrate Polymers,
92: 2224– 2231.

204.

Sun X., Xu F., Zhao H., Sun R., Fowler P. and M.S. Baird. 2005. Physicochemical
characterisation of residual hemicelluloses isolated with cyanamide-activated hydrogen
peroxide from organosolv pre-treated wheat straw. Bioresource Technology, 96: 1342–
1349.

205.

TAPPI. 2006. Tensile properties of paper and paperboard. Revision of T 494 om-01.

206.

Teleman A., Lundqvist J., Tjerneld F., Stålbrand H. and O. Dahlman. 2000.
Characterization of acetylated 4-O-methylglucuronoxylan isolated from aspen employing
1H and 13C NMR spectroscopy. Carbohydrate Research, 329: 807-815.

207.

Thuault A., Eve S., Blond D., Bréard J. and M., Comina. 2014. Effects of the
hygrothermal environment on the mechanical properties of flax fibres. Journal of
Composite Materials, 48: 1699-1707.

208.

Tian M., Wen J., MacDonald D., Asmussen R.M. and A. Chen. 2010. A novel approach
for lignin modification and degradation. Electrochemistry Communications, 12: 527–530.

209.

Timell, T. E.. 1967. Recent progress in the chemistry of wood hemicellulose. Wood
Science and Technology, 1: 45-70.

210.

Toğrul H. and N. Arslan. 2003. Flow properties of sugar beet pulp cellulose and intrinsic
viscosity–molecular weight relationship. Carbohydrate Polymers, 54: 63–71.

211.

Tomé L.C., Gonalves C.M.B., Boaventura M., Brand ão L., Mendes A.M., Silvestre
A.J.D., Neto C.P., Gandini A., Freire C.S.R. and I.M. Marrucho. 2011. Preparation and
evaluation of the barrier properties of cellophane membranes modified with fatty acids.
Carbohydrate Polymer, 83: 836-842.

152

212.

USDA. 1971. Water soluble edible feed packaging film of high amylose starch acetate.
U.S. Patent No. 3553196.

213.

Vartiainen J. and A. Harlin. 2011. Crosslinking as an Efficient Tool for Decreasing
Moisture Sensitivity of Biobased Nanocomposite Films. Materials Sciences and
Applications, 2: 346-354.

214.

Wang S., Wang K., Liu Q., Gu Y., Luo Z. and K. Cen. 2009. Comparison of the pyrolysis
behavior of lignins from different tree species. Biotechnology Advances, 27: 562–567.

215.

Wardrop A. B.. 1969. The structure of the cell wall in lignified collenchyma of Eryngium
sp.(Umbelliferae). Austraalian Journal of Botany, 17 (2): 229-240.

216.

Whistler R.L. and J.R. Daniel. 1984. Molecular structure of starch. In: Whistler R.L.,
Bemiller J.N. and E.F. Paschall (Eds.). Starch: Chemistry and Technology, 2nd Ed..
Academic Press, San Diego, CA. 153-182.

217.

Willett J.L. and R.L. Shrongen. 2002. Processing and properties of extruded
starch/polymer foams. Polymer, 43: 5935-5947.

218.

Wise L. E. and E. C. Jahn, 1952. Wood Chemistry. Reinhold Publishing Co., New York,
NY, USA.

219.

Wyser Y., Pelletier C. and J. Lange. 2001. Predicting and determining the bending
stiffness of thin films andlaminates. Packaging technology and science. 14: 97-108.

220.

Xia Z.Y.,Yoshida T. and M. Funaoka. 2003. Enzymatic degradation of highly phenolic
lignin based Polymers (lignophenols). European Polymer Journal, 39: 909-914.

221.

Xiang Z., Watson J., Tobimatsu Y. and T. Runge. 2014. Film-forming polymers from
distillers’ grains: structural and material properties. Industrial Crops and Products, 59:
282-289.

222.

Yamagaki T., Maeda M., Kanazawa K., Ishizuka Y. and H. Nakanishi. 1997. Structural
clarification of Caulerpa cell wall β-(1→3)-xylan by NMR spectroscopy. Bioscience,
Biotechnology, and Biochemistry, 61: 1077-1080.

223.

Zhang P. Y. and R. L. 2004. Whistler. Mechanical Properties and Water Vapor
Permeability of Thin Film from Corn Hull Arabinoxylan. Journal of Applied Polymer
Science, 93: 2896–2902.

224.

Zhang S., Sun D., Fu Y.Q., and H.J. Du. 2005. Toughness measurement of thin films: a
critical review. Surface & Coatings Technology, 198: 74-84.

225.

Zhang Y., Pitkänenc L., Dougladed J., Tenkanenc M., Remonda C. and C. Joly. 2011.
Wheat bran arabinoxylans: Chemical structure and film properties of three isolated
fractions. Carbohydrate Polymers, 86: 852- 859

153

226.

Zhao Y. and Y. Deng. 2006. Improvement of Peroxide Bleaching Yield and Efficiency of
TMP Using Glyoxal Crosslink Agents. Industrial & Engineering Chemistry Research, 45:
5813-5818.

154