Content uploaded by John Summerscales
Author content
All content in this area was uploaded by John Summerscales on Dec 17, 2018
Content may be subject to copyright.
Accepted for publication in Industrial Crops and Products on 13 July 2017.
1
Targeted pre-treatment of hemp bast fibres for
1
optimal performance in biocomposite materials:
2
A review
3
4
Ming Liua, Anders Thygesena, John Summerscalesb, Anne S. Meyera,*
5
6
a Center for Bioprocess Engineering, Department of Chemical and Biochemical Engineering, Building 229, Technical
7
University of Denmark, 2800 Kongens Lyngby, Denmark.
8
b Composites Engineering, School of Engineering, Plymouth University, Drake Circus, Plymouth, Devon PL4 8AA,
9
United Kingdom
10
11
12
13
Accepted for publication in Industrial Crops and Products on 13 July 2017.
2
Abstract
14
Global interest in the use of plant fibres in natural fibre reinforced composites (NFCs) is growing rapidly. The increased
15
interest is primarily due to the advantageous properties of natural fibres including biodegradability, low cost, low
16
density and high stiffness and strength to weight ratio. In order to achieve strong NFCs, well separated and cellulose-
17
rich fibres are required. Hemp is taking a center stage in this regard as a source of suitable natural plant cellulose fibres
18
because natural hemp bast fibres are long and inherently possess high strength. Classical field and water retting methods
19
have been used for centuries for removal of non-cellulosic components from fibrous plant stems including from hemp,
20
but carries a risk of reducing the mechanical properties of the fibres via damaging the cellulose. For NFCs new targeted
21
fibre pre-treatment methods are needed to selectively and effectively remove non-cellulosic components from the plant
22
fibres and for producing cellulose rich fibres without introducing any damage to the fibres. A key feature for successful
23
use of natural fibres such as hemp fibres in composite materials is optimal interfacial contact between the fibres and the
24
hydrophobic composite matrix material. Targeted modification of natural fibres for NFCs must also be targeted to
25
optimize the fibre surface properties. Consequently, improved interfacial bonding between fibres and hydrophobic
26
polymers, reduced moisture uptake, increased microbial degradation resistance, and prolonged durability of NFCs can
27
be achieved. This review, using hemp bast fibres as an example, critically and comprehensively assesses the targeted
28
pretreatment technologies and data available for producing well separated cellulose bast fibres having optimal chemical
29
and physical properties for maximizing the mechanical performance and durability of NFCs.
30
31
32
Keywords: Hemp fibres; mechanical properties; stiffness; tensile strength; natural-fibre composites
33
Accepted for publication in Industrial Crops and Products on 13 July 2017.
3
1. Introduction
34
Natural plant fibres such as hemp and flax fibres are currently receiving high research attention for use as reinforcing
35
agents in composites to substitute for synthetic fibres (Fuqua et al., 2012; Joshi et al., 2004; Liu et al., 2016; K. L.
36
Pickering et al., 2007; Van Vuure et al., 2015). Hemp (Cannabis sativa L.), known as industrial hemp, is one of the
37
world’s oldest cultivated and most widely used industrial crops which has been cultivated for obtaining long and strong
38
bast fibres. Hemp cultivation moreover allows a diversification of crop rotations in arable farming and has a low
39
requirement for fertilizers and herbicides (Ranalli and Venturi 2004; Amaducci et al. 2014). The flowering top and to a
40
lesser extent the leaves of hemp plants can produce resin secretions containing the narcotic 9-∆ tetrahydrocannabinol
41
(THC), the psychoactive ingredient in marijuana and hashish (Shahzad 2011). However, industrial hemp varieties
42
grown as plant fibre sources produce less than 0.3% dry weight of THC, which is too low to be directly used as a
43
narcotic (Salentijn et al. 2015). The hemp bast fibres have long been used in textile production such as hemp garments
44
and soft furnishings. The co-product, hemp seeds, is a source of more than 10 fatty acids which yield oil of up to 26 −
45
37% by weight (Da Porto et al. 2015).
46
Hemp bast fibres are defined as the continuum of primary and secondary cell walls of the cells that form the cortex
47
sclerenchyma layer of the hemp stem. These bast fibres are particularly long and contain highly crystalline cellulose
48
fibrils. These characteristics make hemp plants a promising source of natural cellulosic fibres (Liu et al. 2015a),
49
including natural fibres for use in biocomposite materials. More importantly, compared to synthetic materials (e.g. glass
50
fibre), hemp fibres also have other advantages such as low cost and low density together with their high stiffness- and
51
strength- to-weight ratios (Faruk et al., 2012; Joshi et al., 2004; Liu et al., 2015; Reddy and Yang, 2008). However, their
52
inherent disadvantages such as a hydrophilic surface, high moisture uptake and vulnerability to microbial attack
53
challenge the direct application of the natural fibres as reinforcement in natural fibre reinforced composites (NFCs).
54
Hydrophilic surfaces will cause weak bonding between fibres and matrix composite materials, particularly hydrophobic
55
matrix polymers. Moisture uptake can thus induce dimensional changes (or swelling), decrease mechanical performance
56
and confer higher susceptibility to microbiological attack. Targeted fibre pre-treatment is therefore important for tuning
57
the properties of the natural fibres for optimal inclusion, mechanical performance, and durability for successful
58
application in NFCs.
59
Accepted for publication in Industrial Crops and Products on 13 July 2017.
4
2. Natural fibre reinforced composites (NFCs)
60
A composite is considered to be any multiphase material that exhibits a significant proportion of the properties of the
61
constituent phases in such way that a better combination of properties is realized (Callister, 1994). Many composite
62
materials are composed of two phases: a continuous phase (the matrix) which surrounds the dispersed or discontinuous
63
second phase (the reinforcement). In natural fibre reinforced composites (NFCs) natural plant fibres constitute the
64
reinformement phase, whereas the matrix can be any (synthetic) polymeric material. In principle, the mechanical
65
properties of a composite material thus depend not only on the fibres, but also on the degree to which an applied load is
66
transmitted to the fibres by the matrix polymers under stress, and thus obviously vary depending on the nature of the
67
matrix material as well (Callister 1994). Figure 1 illustrates several important parameters that influence the mechanical
68
properties of composites, i.e. NFCs. These factors include: (a) fibre properties; (b) matrix properties; (c) fibre length;
69
(d) fibre packing ability; (e) fibre orientation; (f) porosity; (g) fibre volume content; (h) fibre/matrix interface properties
70
(Pickering et al., 2015). Based on these parameters, there is a range of desired properties required to obtain
71
unidirectional composites with high mechanical performance and long durability in practical use (Table 1), and there
72
are both advantages and disadvantages of using natural plant fibres in NFCs (Table 2).
73
Natural fibres are mainly composed of cellulose and other non-cellulosic cell wall components, notably of
74
hemicellulose and aromatic substances. These components provides opportunity to tailor and modify natural fibres with
75
chemical agents and enzymes (Liu et al. 2016b; Liu et al. 2016c). Natural fibre treatments for NFCs are primarily
76
targeted to remove non-cellulosic cell wall components to obtain well separated and cellulose rich fibres, but more
77
recently, treatments are also designed to increase surface hydrophobicity, moisture resistance and microbiological decay
78
resistance of fibres (Bessadok et al. 2007; Özmen 2012; Zhang and He 2013). A primary restriction for successful use
79
of natural fibres in NFCs is the hydrophilic nature of the cellulose fibres, which, as already mentioned briefly above,
80
may result in moisture sensitivity and low microbiological decay resistance of NFCs (Summerscales et al. 2010a)
81
(Table 2). The hydrophobic nature of cellulose fibres, on the other hand, provides the possibility for chemical and
82
enzymatic modification of the hydroxyl groups on fibre surfaces Generally, the purpose of targeted fibre treatment is to
83
overcome the disadvantages of NFCs to obtain: (a) improved bonding between natural fibres and matrix polymers;
84
(b) reduced moisture uptake/increased moisture resistance; (c) improved mechanical properties of natural fibres or
85
avoidance of severe decrease in mechanical properties of natural fibres; (d) improved other properties of natural fibres
86
or NFCs (e.g. thermal stability; anti-microbial degradation etc.) and thus in general combat the potential disadvantages
87
of natural fibres versus synthetic fibers for use in NFCs (Table 2).
88
Accepted for publication in Industrial Crops and Products on 13 July 2017.
5
3. Plant fibre selection
89
3.1. Hemp fibres
90
Plant fibres, the most widely used natural fibres in composites, can be classified into four categories according to
91
their tissue origin in the plant, i.e. whether they come from 1) seed hair, examples are cotton and kapok; 2) leaves or
92
fruit envelopes, as e.g. sisal and coconut; 3) the stem (or bast), e.g. as is the case for fibres from hemp, flax and ramie;
93
or from the 4) root, where examples include sugar beet and carrot (Figure 2) (Vaca-Garcia 2008; Summerscales and
94
Grove 2014). According to current advances in NFCs, the majority of plant fibres currently considered as possible
95
reinforcements for NFCs are bast fibres (Summerscales et al. 2010b; Mehmood and Madsen 2012; Pickering et al.
96
2015; Liu et al. 2016b; Liu et al. 2016c). One reason for this could be that bast fibres, notably bast fibres from hemp and
97
flax, exhibit higher mechanical performances, e.g. strength of 400 – 1000 MPa and stiffness of 25−35 GPa, than fibres
98
from other plant sources and categories, e.g. for cotton the strength is 290 – 600 MPa and stiffness is 5 – 13 GPa
99
(Cierpucha et al. 2004; Vaca-Garcia 2008; Pil et al. 2016). Among commonly used bast fibres for NFCs, hemp bast
100
fibres seem to be the most advantageous also because a) hemp is an extremely fast growing crop, yielding more fibres
101
per acre than other crops such as kenaf (Herer 1985) and flax (van der Werf and Turunen 2008); b) hemp suppresses
102
weeds effectively and is usually free from diseases or pests (van der Werf et al. 1996), and thus c) leave the soil in
103
excellent condition for the succeeding crops, especially when weeds may be troublesome, whereas (non-organic) flax
104
requires high agro-chemical inputs.
105
3.2. Hemp fibre structure
106
(1) At bast fibre level
107
Hemp stems are usually 1.5 – 2.5 m tall and 5 – 15 mm in diameter. The stem contains 30 – 40% (w/w) bast fibres
108
and is organized in layers comprising from the pith to the surface 1 – 5 mm xylem, 10 – 50 µm cambium, 100 – 300 µm
109
cortex, 20 – 100 µm epidermis and 2 −5 µm cuticle (Liu et al. 2015a) (Figure 3). Hemp bast fibres contain primary and
110
secondary fibres situated in the cortex. The primary fibres are more suitable for NFCs because secondary fibres are
111
shorter (approx. 2 mm long) and thinner (approx. 15 µm in diameter) than primary fibres (approx. 20 mm long and 10 –
112
40 µm in diameter) (van der Werf et al. 1994).
113
(2) At single fibre level
114
Similar to other plant fibres, hemp fibre cell walls are mainly composed of a thin primary wall (70 – 110 nm) and a
115
thick secondary cell wall (3 – 13 µm) (Thygesen et al. 2006). The secondary cell wall is composed of a S1 layer (100 –
116
Accepted for publication in Industrial Crops and Products on 13 July 2017.
6
130 nm in thickness) and a S2 layer (3 – 13 µm in thickness) (Thygesen et al. 2006). The thicker S2 layer has been
117
observed to have a laminate structure consisting of 1 – 4 major concentric layers with a thickness of 1 – 5 µm (Figure
118
4) (Thygesen et al. 2006). The construction varies greatly between the layers:
119
the primary wall has loosely packed microfibrils which interweave randomly;
120
the secondary walls have microfibrils closely packed and parallel to each other (Thygesen et al. 2006; Fan
121
et al. 2011). Using X-ray diffraction, the cellulose microfibrils in the S1 layer have an S-helical orientation
122
with microfibril angles (MFA) in the range of 70 – 90°, while in the S2 layer, the microfibrils have Z-helical
123
orientation with MFA in the range of 25 – 30° (outer part) and 0 − 2° (inner part) (Thygesen et al. 2006).
124
(3) At microfibril level
125
The fibre cell wall can in itself be considered as a composite (Figure 4) composed of cellulose microfibrils as
126
reinforcements and non-cellulosic polymers, mainly including hemicellulose, pectin, lignin, as matrix polymers
127
(Lefeuvre et al. 2014). Cellulose microfibrils are cross-linked by glycans (e.g. xyloglucan, glucuronoarabinoxylan,
128
galactomannan, mannan) (Carpita and Gibeaut 1993; Peña et al. 2008). The interlocked network of microfibrils and
129
glycans is further embedded in a matrix of pectic substances and reinforced with structural aromatic substances (e.g.
130
lignin and hydroxycinnamates) (Markwalder and Neukom 1976; Chesson et al. 1983; Carpita and Gibeaut 1993). The
131
aromatic substances presumably add mechanical properties to the cell by interacting with polysaccharides via cross-
132
linking reactions (Ralph 2010; Lupoi et al. 2015) and are associated with ageing and stiffening of the plant.
133
3.3. Chemical composition
134
Hemp fibres, similarly to other plant fibres, essentially contain 5 major components (Crônier et al. 2005; Liu et al.
135
2015a; Liu et al. 2015b): (1) structural polysaccharides: cellulose and hemicellulose (the hemicellulose mainly being
136
xyloglucan); (2) structural protein; (3) other polysaccharides, notably pectin (homogalacturonan); (4) lignin; (5) waxes,
137
and (6) minerals. Hemp fibres are generally composed of 53-91% cellulose, 4-18% hemicellulose, 1-17% pectin, and 1-
138
21% lignin (Table 3) , but the chemical composition of untreated hemp bast fibres varies with the cultivar (see Table
139
3), harvest year (Jankauskiene et al. 2015), harvest time (or growing stage) (Crônier et al. 2005), the location of fibres
140
within the stems (Charlet et al. 2007), and the final composition also depends on the type of fibre processing (Korte and
141
Staiger 2008; Nykter et al. 2008). The chemical composition of fibres and the distribution of the constituents define the
142
properties of the fibres (Komuraiah et al. 2014). Changes in chemical composition of natural fibres after different fibre
143
processing regimes have been commonly used to explain changes in the mechanical properties of fibres and fibre/matrix
144
polymers composites (Charlet et al. 2007; Thuault et al. 2013; Behazin et al. 2016).
145
Accepted for publication in Industrial Crops and Products on 13 July 2017.
7
3.4. Mechanical properties
146
The density of hemp fibres is 1.4-1.6 g/cm3, whereas the mechanical properties of fibres from different hemp
147
varieties, i.e. tensile strength, stiffness and failure strain range from 200-1000 MPa, 18-66 GPa, and 2-4%, respectively
148
(Table 4). In comparison, synthetic fibres including glass fibres and carbon fibres have relatively high tensile strengths
149
of 2000 and up to 4000 MPa, respectively, as well as higher average stiffness of 80 and 238 GPa, respectively (Table
150
4). Glass fibers have much higher density (2.55 g/cm3) than hemp fibers, however, but carbon fibres have low density
151
(1.3 g/cm3) (Table 4). Thus, the specific tensile strength (1.3- 6.7 × 105 m2/s2) and stiffness (1.2-4.4 × 107 m2/s2) of
152
hemp fibres are comparable to the specific tensile strength (approx. 7.6 × 105 m2/s2) and stiffness (approx. 3.0 × 107
153
m2/s2) of glass fibres. Hemp fibres therefore have potential as replacements for glass fibre as reinforcements for
154
composite materials.
155
However, as evident when comparing the available data for hemp fibres (Table 4) there is large variability in the
156
mechanical properties of hemp fibres. Many efforts have been made to understand the properties determining the
157
mechanical properties of hemp fibres (Charlet et al. 2007; Marrot et al. 2013). A number of studies indicate that many
158
inherent factors may affect the mechanical properties of natural fibres, including notably the chemical composition
159
(Charlet et al. 2007; Bourmaud et al. 2013; Marrot et al. 2013), microfibril angle (Baley 2002; Neagu 2005; Nilsson and
160
Gustafsson 2007; Dai 2010; Bourmaud et al. 2013), structure of fibre cell walls (Carpita and Gibeaut 1993; Álvarez et
161
al. 2011; Marrot et al. 2013), in addition to damage incurred during growth and processing. Even for untreated hemp
162
fibres, there is a large scatter in their mechanical properties. In addition to the factors that affect the chemical
163
composition, the factors affecting the mechanical properties of untreated hemp bast fibres include variety (Marrot et al.
164
2013), growth stage of the plant at harvest (Keller et al. 2001; Liu et al. 2015a), growth conditions (van der Werf et al.
165
1995), stem section (Duval et al. 2011; Liu et al. 2015b).
166
Marrot et al. (2013) compared the mechanical properties of hemp fibres from two different varieties of hemp plants,
167
i.e. Fedora 17 and Felina 32 and found that the two types of fibres exhibited different tensile strength, stiffness and
168
strain, which was attributed to the differences in chemical composition especially matrix polymers e.g. pectin between
169
the two types of investigated fibres. Similar results were reported for flax fibres (Bourmaud et al. 2013). The growth
170
stage of the hemp plant has also been found to influence the mechanical properties of the fibres (Keller et al. 2001; Liu
171
et al. 2015a). Fibres from hemp plants harvested at the beginning of flowering thus had higher tensile strength (i.e. 950
172
MPa) and stiffness (i.e. 35GPa) than fibres from hemp plant harvested at seed maturity with a tensile strength of 810
173
MPa and stiffness of 31 GPa (Liu et al. 2015a). The low mechanical performance of fibres from hemp plants harvested
174
Accepted for publication in Industrial Crops and Products on 13 July 2017.
8
at seed maturity are suggested to be mainly due to low cellulose and partly low pectin content and to formation of
175
secondary fibres which are shorter and have less favorable mechanical properties than primary fibres (Liu et al. 2015a).
176
Apart from the factors mentioned above, the location of the fibres in hemp stems also influences the mechanical
177
properties of the hemp fibres (Duval et al. 2011; Liu et al. 2015b). Hence, fibres from the middle section of the hemp
178
stem had highest mechanical properties followed by fibres from the top and bottom sections. The differences in
179
mechanical properties of hemp fibres from different stem sections are attributed to the differences in their chemical
180
composition particularly the cellulose and pectin content, and to the proportion of secondary fibres (Liu et al. 2015b).
181
4. Matrix polymer
182
For a fibre reinforced polymer composite, stiffness and strength of the matrix polymer are lower than that of the
183
reinforcements. Matrix polymer, an important part of the NFCs, protects reinforcements against adverse environments,
184
mechanical abrasion, and transfers stresses to the different forms of reinforcements (e.g. fibres or fibre fabrics) in
185
different shapes to meet practical needs (Pickering et al. 2015). Natural fibres are sensitive to water and high
186
temperature. Therefore, the most commonly used matrix polymers in NFCs are those that can be processed (i.e. melted
187
or cured) at low temperature (e.g. below 200 °C) and which have low water content. Those polymers can be categorized
188
as thermoset and thermoplastic polymers, respectively.
189
A large range of thermoplastic polymers (Table 5) and thermoset polymers (Table 6) have been applied for NFCs.
190
These polymers commonly have a density of 0.9 – 1.6 g/cm3, and stiffness of 0.2 – 14 GPa. As mentioned, the melting
191
point as well as the curing temperature of commonly used thermoplastic and thermoset polymers is below 200 °C. In
192
practice, this relatively low temperature required for the processing, means that manufacturing of NFCs with such
193
polymers does not significantly influence the mechanical properties of the resultant NFCs. However, the glass transition
194
temperature (Tg) is normally 200 ± 50 °C below the crystalline melting point (Tm), so a thermoplastic matrix system that
195
can be processed without degrading the fibres is unlikely to be useable when creep deformation may occur. Exceptions
196
include poly (lactic acid) (PLA) or thermosetting resins where the Tg is close to the maximum cure temperature.
197
5. Pre-treatment of hemp fibres affecting mechanical performance of NFCs
198
As already stressed in Section 2, above, the hydrophilic character of natural fibres is generally considered a major
199
barrier for use of natural fibres in composites where high reliability and stability of fibre properties are required. The
200
hydrophilic character of natural fibres can cause the following issues in NFCs: 1) Decreased interfacial contact between
201
the natural fibre and the matrix polymer; 2) water uptake which in turn may decrease stability and durability of the
202
Accepted for publication in Industrial Crops and Products on 13 July 2017.
9
NFCs, and 3) provide conditions for microorganisms to thrive. The pretreatment or processing of the natural fibres may
203
however be designed to help alleviate the hydrophilicity and even provide for enhancing the natural properties of the
204
fibres for successful use in NFCs. It is however important to understand the effects of fibre pre-treatments on the
205
mechanical properties of NFCs in order to provide optimal cost-benefit fibre treatments.
206
5.1. Defibration
207
The defibration process, a term commonly used for the pulping processes for paper, involves breakdown of the wood
208
matrix and results in separation of the fibres. In relation to fibre processing for use in NFCs, the defibration process
209
refers to the treatments applied to degrade non-cellulosic components of the natural fibres to produce small fibre
210
bundles (technical fibres) or single fibres (ultimate fibres) from large fibre bundles. Several fibre treatment methods,
211
including traditional retting, steam explosion, chemical treatment, controlled fungal retting and enzyme treatment etc.,
212
have been tested to defibrate natural fibres for composites use. Optimal defibration of natural fibres produces increased
213
surface area for contacting with matrix polymers and allow for the matrix polymers to fill the space between fibres or
214
fibre bundles (Liu et al. 2016b). Interfacial contact, i.e. the bonding or attachment, between the fibres and the matrix
215
polymer is therefore increased. Furthermore, defibration of natural fibres produces cellulose-rich fibres by removal of
216
non-fibre cells that have large voids (i.e. parenchymal and epidermal cells), and this may in turn decrease the porosity
217
of the resultant composites. Consequently, the mechanical properties of NFCs can be increased by correct defibration.
218
5.1.1. Traditional retting
219
Retting is a process resting on the (random) action of microorganisms to remove non-cellulosic components from
220
natural fibres and separate fibres from the plant stem structure so as to obtain cellulose-rich fibres (Thygesen et al.
221
2013). Two retting methods have been traditionally used: field retting (also known as dew retting), and water retting.
222
In field retting, plant stems are cut and spread out in the fields where they are casually attacked by microbial
223
activities (Liu et al. 2015a). The growth of microbial communities, especially the pectinolytic microbial communities,
224
largely depends on the local microbiota, the moisture content of plant stems, and on the humidity and temperature under
225
local weather conditions. Therefore field retting is subject to geographical variation and depend on whether the weather
226
conditions are suitable for microbial proliferation; the retting processing usually takes several weeks (Jankauskienė and
227
Gruzdevienė 2013; Liu et al. 2015a). Water retting is performed by submerging plant stems in water in rivers, lakes or
228
tanks where pectinolytic microbial communities develop. The water can penetrate into the plant stem structures,
229
increase the moisture absorption of plant samples, and boost the proliferation of microorganisms. Usually, water retting
230
takes less time e.g. 1-2 weeks than field retting (Thygesen et al. 2013; Jankauskienė and Gruzdevienė 2013).
231
Accepted for publication in Industrial Crops and Products on 13 July 2017.
10
We have studied the effect of field retting on chemical composition and mechanical properties of hemp fibres and
232
hemp fibre/epoxy composites (Liu et al. 2015a; Liu et al. 2016d) and found that both fungal and bacterial communities
233
were involved in field retting. Species of the Ascomycota phylum were observed during the first 1 – 2 weeks of field
234
retting, after which different types of bacteria, notably different Proteobacteria proliferated during the 2nd to the 3rd
235
week of retting (Liu et al. 2016d). Different enzyme activities were observed in field retted samples including
236
glucanase, polygalacturonase, galactanase, and xyloglucan (XG)-specific endoglucanase (Liu et al. 2016d). A summary
237
of the enzyme activities recorded during field retting are shown in Table 7.
238
In field retting, pectin as assessed by galacturonic acid (polygalacturonan) content is gradually removed during the
239
retting in response to the presence of pectinolytic enzyme activities. The polygalacturonan content of hemp fibres was
240
thus found to decrease to 3 – 4% from 8% after 2 weeks of field retting (Liu et al. 2015a; Liu et al. 2016d). However,
241
the cellulose content of hemp fibres also decreased during field retting, particularly after 2 weeks (Liu et al. 2015a), in
242
accord with the presence of glucanase activity in the field retted samples (Liu et al. 2016d). The stiffness of hemp fibres
243
decreased from 31 GPa to 28 GPa and tensile strength decreased from 810 MPa to 680 MPa after 20 days of field
244
retting (Liu et al. 2015a). This decrease in mechanical properties of fibres was attributed to the loss and damage of the
245
cellulose in the fibres (Liu et al. 2015a).
246
When the effect of field retting and water retting on chemical composition and physical properties of hemp fibres
247
was compared it has been found that water retted hemp fibres had higher content of cellulose (81.7%), hemicellulose
248
(6.3%) and lower content of lignin (10.2%) when compared with that of field retted fibres (78.4%, 5.9% and 13.1%,
249
respectively) (Jankauskienė and Gruzdevienė 2013). Candilo et al. (2010) compared the performance of a traditional
250
water retting process, without inoculum, and a modified water retting process inoculated with two selected pectinolytic
251
bacteria (i.e. an anaerobic strain Clostridium sp. L1/6 and an aerobic strain Bacillus sp. ROO40b) for hemp fibre
252
treatments and fouond that the modified water retting process speeded up the retting process and significantly improved
253
fibre quality (Di Candilo et al. 2010). The best fibre quality with a cellulose content of 89% could be obtained after only
254
3 – 4 days of retting with the addition of the bacterial inoculum (Di Candilo et al. 2010). Thygesen et al. (2013)
255
compared the morphology of cross sections of composites made with water retted hemp fibres and untreated fibres and
256
found that water retting appeared to improve the interface between fibres and polymer matrix (i.e. epoxy) and that water
257
retted fibres reinforced epoxy composites had fewer voids than the benchmark (Thygesen et al. 2013). Generally, it is
258
thus believed that water retting provides higher retting efficiency and stronger fibres than field retting.
259
Accepted for publication in Industrial Crops and Products on 13 July 2017.
11
Undoubtedly, traditional water retting of hemp consumes large amounts of freshwater, and causes eutrophication
260
(van der Werf and Turunen 2008). Efforts have been made to make this treatment less reliant on the availability of
261
water resources by using seawater to substitute fresh water for water retting (Zhang et al. 2008). It was found that
262
seawater works as well as fresh water in retting of hemp fibres with regard to the removal of non-cellulosic
263
components, as pectin content decreased from 6.5% to 0.4% after 2 weeks of seawater retting (Zhang et al. 2008).
264
5.1.2. Steam explosion
265
Steam explosion is a thermo-mechanical-chemical defibration method which allows breakdown of the lignocellulosic
266
structural components via a sharp pressure change. Steam explosion has been widely used as a pre-treatment technology
267
for lignocellulosic materials to improve enzyme catalyzed cellulose degradation (Jacquet et al. 2015), and has also been
268
used for pre-treatment of natural fibres intended for use in composites in order to defibrate fibre bundles into single
269
fibres and small fibre bundles by degrading or disrupting the middle lamella (ML) between the individual fibres
270
(Vignon et al. 1996; Keller 2003; Thomsen et al. 2006; Kukle et al. 2011).
271
Keller (Keller 2003) compared biological processes and steam explosion for pectin degradation on hemp fibres, and
272
studied the mechanical performance of the treated fibres in composites. Hemp fibres produced from steam explosion
273
were four times as long (~ 0.10 mm) as the fibres obtained after biological treatment and steam exploded fibres
274
therefore had a significantly higher aspect ratio (i.e. 23) than biologically treated hemp fibres (i.e. 4). Composites with a
275
fibre volume fraction of up to 42% of steam exploded hemp fibres having a tensile strength of up to 30 MPa and
276
stiffness of up to 4 GPa could be achieved. Steam explosion was also shown to be more effective in degrading pectin
277
from the middle lamella regions, allowing production of small fibre bundles and elementary bast fibres, when steam
278
explosion treatment was combined with biological retting pre-treatment (Vignon et al. 1996). A similar conclusion was
279
drawn from results showing that steam explosion of retted hemp fibres increased the cellulose content of the fibres from
280
73% to 85 – 90%, while that of raw hemp fibres was increased from 60 – 64% to 73 – 75% (Thygesen 2006). Further
281
confirmation was made that combination of retting methods and steam explosion gave good disintegration of hemp
282
fibres, and that alkali pre-treatment after retting is not necessary prior to steam explosion (Kukle et al. 2011).
283
5.1.3. Chemical treatment
284
Chemical treatment involves use of different chemicals to remove non-cellulosic components, i.e. particularly pectin,
285
hemicellulose and lignin, from natural fibres. Chemical chelators such as ethylenediaminetetraacetic acid (EDTA) and
286
ethylene diamine tetra (methylene phosphonic acid) (EDTMPA) have been applied in natural fibre defibration to loosen
287
the structure of pectin (Evans et al. 2002; Stuart et al. 2006; Thygesen et al. 2006; Li and Pickering 2008; Bacci et al.
288
Accepted for publication in Industrial Crops and Products on 13 July 2017.
12
2010; Liu et al. 2016b). Alkalis such as sodium hydroxide (NaOH) and potassium hydroxide (KOH) are commonly
289
used to help remove hemicellulose (e.g. xyloglucan) from natural fibres. Other non-cellulosic components, notably
290
lignin and pectin can also be partially removed by alkali treatments (Korte and Staiger 2008; Sawpan et al. 2011a; Liu
291
et al. 2016b). Sodium sulphite (Na2SO3) is also commonly used together with alkali to remove lignin (Wang et al.
292
2003).
293
Pectin is mainly located in the middle lamella (ML) of hemp fibres where it functions as glue packing the single
294
fibres together. Removal of pectin improves production of small cellulosic fibre bundles or individual fibres. The most
295
abundant pectic polysaccharide in plant cell walls is homogalacturonan (HG). Calcium, as described by the ʺegg-box"
296
model, is important for stabilizing the structure of HG via calcium-mediated interactions (Liners et al. 1989). Chemical
297
chelators such as EDTA and EDTMPA are capable of removing calcium from natural fibres through competitive
298
chelation of calcium (Griko 1999).
299
We have recently studied the effect of different concentrations of EDTA on the removal of calcium and HG from
300
hemp bast fibres (Liu et al. 2016b). Calcium removal from hemp fibres was shown to be positively correlated with the
301
concentration of EDTA used. The calcium released increased from 6 (mg/100 g dry matter) in the control treatment (i.e.
302
0% EDTA) to 490 (mg/100 dry matter) with 0.5% EDTA treatment, and finally to about 800 (mg/100g dry matter) with
303
treatment with 2% or 3% EDTA, respectively (Liu et al. 2016b). The removal of HG from hemp fibres correlated
304
positively and linearly with calcium removal by EDTA with a slope of 1.0 (mole ratio of galacturonic acid and calcium
305
release) (Liu et al. 2016b). With the removal of calcium, more HG was thus removed from hemp bast fibres. After
306
EDTA treatment, slightly increased separation of hemp fibres has been observed in several studies (Le Troedec et al.
307
2008; Li and Pickering 2008; Liu et al. 2016b) and improved mechanical properties of hemp fibre reinforced polymer
308
composites made from EDTA treated hemp fibres have been reported (Li and Pickering 2008). In our own work, we
309
obtained composites with lower porosity and higher mechanical properties with EDTA treated fibres compared to
310
composites made with untreated fibres (Liu et al. 2016b). However, the data have also revealed that pectin can only be
311
partially removed by EDTA treatment, even though most of the calcium in hemp fibres can be removed (Liu et al.
312
2016b). Enhanced pectin removal was achieved by combining chemical chelators and pectinolytic enzymes (i.e.
313
polygalacturonase treatment) (Li and Pickering 2008; Liu et al. 2016b). The combined treatment with chemical
314
chelators and enzymes will be discussed in detail in sections 5.1.4 and 5.1.5.
315
Alkali extraction with NaOH or KOH is widely used for the isolation of hemicellulose from lignocellulosic biomass
316
to obtain cellulose of high purity (Puls et al. 2006). We have previously investigated the effect of removal of
317
Accepted for publication in Industrial Crops and Products on 13 July 2017.
13
hemicellulose with 10% NaOH after pectin was removed by EDTA and pectinase and found that up to 70% of the
318
hemicellulose could be removed from hemp fibres by NaOH (Liu et al. 2016b). This removal of pectin and
319
hemicellulose promoted separation of the hemp fibres and improved the cleanliness of the fibre surface (Liu et al.
320
2016b), as shown in Figure 5. These changes in fibre morphology and degree of fibre separation resulted in
321
significantly increased stiffness of the resultant composites; the improved stiffness was interpreted to be mainly due to
322
the increased cellulose content of the fibres (up to 86%) as well as decreased composite porosity as a result of the
323
improved interface bonding between the fibres and the matrix polymers, as confirmed by ESEM micrographs (Liu et al.
324
2016b). However, the tensile strength of composites with NaOH treated fibres decreased, apparently as a result of
325
disruption of the cellulose-hemicellulose interlocked network after NaOH treatment (Liu et al. 2016b).
326
Treatments combining alkali and sodium sulphite have also been shown to be effective in lignin removal from hemp
327
fibres, and the concentration levels of both NaOH and Na2SO3 are known to significantly affect lignin removal (Wang
328
et al. 2003). Combined treatment with alkali (i.e. 5% NaOH) and sodium sulphite (i.e. 2% Na2SO3) was reported to
329
enhance separation of fibres from fibre bundles, remove lignin and the other components, expose cellulose hydroxyl
330
groups, make the fibre surfaces cleaner and also enhance thermal stability of the fibres by increasing cellulose
331
crystallinity in turn improving the mechanical properties of hemp fibre/epoxy composites (Islam et al. 2011).
332
5.1.4. Controlled microbiological retting
333
Controlled microbiological retting is a process employing controlled incubation of stems or bast fibres with selected
334
microorganisms, which secrete enzymes (notably pectinolytic enzymes) that attack the non-cellulosic components of
335
natural fibres during the retting incubation. Traditional field retting and water retting are not well controlled and depend
336
on both spontaneous proliferation of microbes prevalent in the particular geographic region and on the availability of
337
water, i.e. via atmospheric precipitation. Controlled microbiological retting can overcome some of the disadvantages of
338
traditional retting methods and also greatly speed up the retting process.
339
Thygesen et al. compared the mechanical properties of composites made from controlled fungally defibrated hemp
340
fibres with composites made from traditional water retted hemp fibres (Thygesen et al. 2007). Two strains of white rot
341
fungi, Phlebia radiata Cel 26 (a mutated fungus low in cellulase expression) and Ceriporiopsis subvermispora, were
342
employed in the controlled microbiological (fungal) treatment. Fibres produced via controlled retting with P. radiata
343
Cel 26 had higher cellulose content (78%, w/w) than water retted hemp fibres (74%) and C. subvermispora retted fibres
344
(72%). Hemp fibres retted with P. radiata Cel 26 moreover exhibited the highest effective stiffness and strength in
345
composites (94 GPa and 643 MPa, respectively) compared to water retted fibres (88 GPa and 586 MPa, respectively).
346
Accepted for publication in Industrial Crops and Products on 13 July 2017.
14
The improvement in the mechanical performance of epoxy composites reinforced with P. radiata Cel 26 retted fibres
347
was explained as being due to improved surface contact between the hemp fibres and the epoxy polymer matrix
348
(Thygesen et al. 2007).
349
In further studies of retting of hemp stems from different stem sections (i.e. top, middle and bottom) with P. radiata
350
Cel 26 and C. subvermispora, respectively, it was also found that P. radiata Cel 26 exhibited higher depectinization
351
selectivity and efficiency than C. subvermispora (Thygesen et al. 2013; Liu et al. 2015b). Moreover, fibres retted with
352
P. radiata Cel 26 had better mechanical properties with tensile strength, strain and stiffness of 736 MPa, 2.3% and 42
353
GPa than the fibres treated with C. subvermispora. In a more thorough comparison between fungal retting of hemp
354
fibres with P. radiata Cel 26 and traditional field retting, including examination of differences in enzyme activities and
355
mechanical properties of fibres and fibre reinforced composites resulting from the treatments, a lower β-glucanase
356
activity and higher polygalacturonase activity was demonstrated in P. radiata Cel 26 retted hemp fibres than in field
357
retted samples (Table 6) (Liu et al. 2016d). Composites made with P. radiata Cel 26 retted hemp fibres exhibited
358
significantly higher strength and stiffness than composites made with field retted fibres (Liu et al. 2016d).
359
Li et al. compared the mechanical properties of hemp fibres and their composites after 2 weeks of fungal retting with
360
the white rot fungi Schizophyllum commune (Li et al. 2009). Composites with fibres treated with S. commune had a
361
higher tensile strength of 45 MPa and stiffness of 7 GPa compared to the composites with untreated fibres which had
362
tensile strength of 35 MPa and stiffness of 4 GPa (Li et al. 2009). Results from both single fibre pull-out tests and the
363
Bowyer and Bader model (Li et al. 2009) moreover showed that the interfacial shear strength (IFSS) of S. commune
364
treated fibre composites was higher than that of untreated fibre composites. Based on these data it was suggested that
365
the hemp fibre interfacial bonding with polypropylene matrix was improved by white rot fungi treatment with C.
366
commune. In general, the findings reported by Li et al. are consistent with the studies by Thygesen et al. (Thygesen et
367
al. 2007; Thygesen et al. 2013) and Liu et al. (Liu et al. 2016d).
368
5.1.5. Enzyme treatment
369
Enzymes, as biocatalysts, can catalyze selective removal of non-cellulosic components from hemp fibres. Compared
370
to chemical treatment, biocatalyzed reactions can be performed under mild conditions, i.e. low temperature and near
371
neutral pH. Enzyme treatment involving mainly pectinolytic enzymes thus offers an environmentally friendly as well as
372
an efficient method for separating fibres and removing non-cellulosic components from natural plant fibres. In practice,
373
treatment of natural plant bast fibres with pectinolytic enzymes (e.g. endo-polygalacturonase) induce release of pectic
374
polymers from the ML and fibre cell walls as the enzymes catalyze random hydrolysis of the glycosidic bonds of the
375
Accepted for publication in Industrial Crops and Products on 13 July 2017.
15
HG backbone to liberate monomeric, dimeric or oligomeric fragments (Benen et al. 1999). After removal of pectin from
376
the ML, the bonding between fibres becomes weaker and individual fibres and small fibre bundles can be separated
377
from the larger fibre bundles (Liu et al. 2016b).
378
We recently employed 0.2% of a mono-active fungally derived endo-polygalacturonase to treat hemp fibres to
379
investigate the effect of the treatment on fibre and fibre reinforced composite properties (Liu et al. 2016b). After endo-
380
polygalacturonase treatment, the polygalacturonan content decreased to 4.8% from 8.3% in untreated hemp fibres (Liu
381
et al. 2016b). The fibre surfaces became cleaner after the treatment and improved mechanical properties of fibre
382
reinforced composites were obtained compared to the composites produced with traditional field retted fibres (Liu et al.
383
2016b). The pectin removal results were in accord with those of (Li and Pickering 2008) who reported a decrease in
384
pectin content from 6.2% in untreated fibres to 5.7% after hemp fibres were treated with pectinase. However, the
385
enzymes cannot directly penetrate the hemp fibres to degrade pectins efficiently. The accessibility of the substrate
386
surface to enzymes is therefore important in enzyme treatment of natural fibres, and pre-treatment of natural fibres is
387
thus necessary to achieve well separated fibres with pectinase treatment.
388
As mentioned above, addition of chemical chelators (e.g. EDTA) has been shown to promote enzyme catalyzed
389
degradation of HG from cellulosic fibres during pectinase treatment (Adamsen et al. 2002; Stuart et al. 2006). As
390
mentioned, the enhanced enzymatic degradation of HG results from the capacity of chemical chelators to competitively
391
form complexes with divalent ions, in turn removing e.g. calcium from pectin (Griko 1999). We found that once 0.5%
392
EDTA was added to 0.2% endo-polygalacturonase during hemp fibre treatment, a significantly higher amount of pectin
393
was removed, leaving only 2% polygalacturonan left in the fibres, compared to 5% for fibres treated with enzyme alone
394
(Liu et al. 2016b). This enhanced pectin removal produced better separation of hemp fibres from fibre bundles and also
395
improved the mechanical properties of the hemp fibre/epoxy composites (Liu et al. 2016b). Enhanced enzymatic
396
removal of pectin and improved mechanical properties of fibre reinforced composites after hemp fibres were treated
397
with chemical chelators and pectinases have also been reported in other studies (Li and Pickering 2008).
398
Besides use of chemical chelators, hydrothermal pre-treatment was found to improve enzymatic removal of pectin
399
from hemp fibres (Liu et al. 2016c). A systematic investigation of hydrothermal pre-treatment at different severities (i.e.
400
30 min at 112 °C, 121 °C and 134 °C, respectively) to promote enzymatic treatment of hemp fibres with pectinases
401
showed that mild hydrothermal pre-treatment could impart better enzymatic removal of pectin from the hemp fibres to
402
produce cellulose rich fibres without damaging the fibre properties (Liu et al. 2016c). It was also found that
403
hydrothermal pre-treatment at 121 °C for 30 min followed by a 90 min pectinolytic enzyme treatment resulted in hemp
404
Accepted for publication in Industrial Crops and Products on 13 July 2017.
16
fibres with low pectin content of 3%, high tensile strength of 780 MPa and stiffness of 36 GPa; fibres treated using this
405
method could be used to make fibre/epoxy composites with tensile strength of 325 MPa and stiffness of 38 GPa at a
406
fibre volume content of 50% (Liu et al. 2016c).
407
5.1.6. Correlation between fibre defibration and mechanical properties NFCs
408
As summarized above, increased tensile strength and stiffness of NFCs have been observed after fibres were pre-
409
treated using different fibre pre-treatment methods. Fibre pre-treatment primarily affects chemical composition of fibres
410
by removing non-cellulosic components (e.g. pectin and hemicellulose), and produces well separated fibres. Good
411
separation of fibres in turn improves interfacial bonding between fibres and matrix polymers. As a result, increased
412
tensile strength and stiffness of NFCs are achieved. Recently, the correlation between different fibre pre-treatment
413
methods and mechanical performance of NFCs with differently treated hemp fibres was established by Liu et al.
414
(2016b). The correlation was established between chemical composition of hemp fibres and porosity (as indicated by
415
fibre correlated porosity factor) of hemp fibre reinforced composites (Liu et al. 2016b), as shown in Figure 6. It was
416
found that less pectin and hemicellulose content of fibres after different treatments resulted in lower porosity of NFCs
417
(Figure 6a). The decrease in porosity of NFCs was explained by changes in fibre microstructure (e.g. fibre separation)
418
and cleanliness of fibre surface after fibre treatment (Liu et al. 2016b). Consequently, increased tensile strength and
419
stiffness of NFCs were obtained via decreasing the porosity of NFCs: The lower the NFC porosity is, the higher is the
420
mechanical performance (Figure 6b) (Liu et al. 2016b). Traditional field retting does not fit such correlation (Figure 6),
421
which may be due to degradation of cellulose caused by field retting (Liu et al. 2016b).
422
5.2. Polymerization/Cross-linking
423
Natural fibres can be considered as a composites with cellulose reinforcement in non-cellulosic (i.e. pectin and
424
lignin) matrix polymers. Pectin and lignin are mainly located in the ML regions between fibres. Lignin in hemp fibres
425
accounts for 2 – 5% of dry matter of the fibre cell walls. Previous studies have shown that model lignin compounds can
426
be catalyzed by laccases or peroxidases to form complex structures, and laccases can also catalyze polymerization of
427
different types of lignin to different degrees (Ikeda et al. 1996; Mattinen et al. 2008). The polymerization of lignin and
428
other aromatic substances has been suggested to improve the mechanical properties of plant cell walls (Ralph 2010).
429
We recently studied the effect of laccase treatment of hemp fibres on properties of fibres and their composites (Liu et
430
al. 2016a). After the hemp fibres were pre-treated with 0.5% EDTA and 0.2% endo-polygalacturonase (EPG), laccase
431
treatment was found to increase stiffness of the resulting fibres and unidirectional hemp fibre/epoxy composites. The
432
resultant composites had stiffness of up to 42 GPa and tensile strength of up to 326 MPa at a fibre volume content of
433
Accepted for publication in Industrial Crops and Products on 13 July 2017.
17
50% (Liu et al. 2016a). The thermal resistance of hemp fibres also increased after laccase treatments, and the maximum
434
degradation temperature increased by about 5 °C. These changes in the mechanical properties of hemp fibres and their
435
composites were attributed to polymerization of lignin by laccase (Liu et al. 2016a).
436
Improvement in the mechanical properties of natural fibres after laccase treatments was also observed for plantain
437
fibres (Álvarez et al. 2011). An increase in the maximum degradation temperature and mechanical properties of treated
438
fibres with increasing concentration of laccase used in the treatment was observed (Álvarez et al. 2011). The maximum
439
degradation temperature increased from 327 °C for the control to 328 °C for fibres treated with laccase at a
440
concentration of 6 U/g dry matter, to 329 °C for fibres treated with laccase at a concentration of 12 U/g dry matter, and
441
finally to 331 °C for fibres treated with laccase at a concentration of 24 U/g dry matter; for the same series of laccase
442
concentration treatments, stiffness of the resulting fibres increased from 1.3 GPa, to 3.5 GPa, to 3.6 GPa, and finally to
443
3.6 GPa, respectively and tensile strength increased from 13.3 MPa, to 17.5 MPa, to 18.6 MPa, and finally to 18.7
444
MPa, respectively (Álvarez et al. 2011).
445
5.3. Surface modification
446
The hydrophilic nature of cellulose rich fibres is a restriction for successful use of natural fibres in composite
447
applications. The hydrophilicity of natural fibres is mainly contributed by hydroxyl groups of cellulose and other
448
hydrophilic non-cellulosic components. Surface modification aims to reduce the hydrophilic nature of cellulose rich
449
fibres mainly by replacing hydroxyl groups of cellulose with less hydrophilic (e.g. O-Na+) or hydrophobic chemical
450
groups (e.g. acetyl groups (CH3CO)). Reduction of the hydrophilic nature of cellulose rich fibres is thought to increase
451
bonding between fibres and hydrophobic polymers. The increased bonding in turn improves the mechanical and
452
physical properties of NFCs.
453
5.3.1. Alkali treatment
454
Besides the removal of a certain portion of hemicellulose and lignin, alkali (NaOH) treatment can reduce the
455
amorphous hydroxyl group by the reaction between alkali and hydroxyl groups, as shown below. (Kabir et al. 2013).
456
The removal of hemicellulose and lignin covering materials can expose more hydroxyl groups of cellulose to alkali. As
457
a result, the hydrophilic nature of natural fibres is reduced and the surface of natural fibres become very clean and
458
smooth (Sawpan et al. 2011a; Liu et al. 2016b), and those changes can improve adhesion between fibres and matrix
459
binders. The effect of alkali treatment on fibre chemical composition, mechanical properties of fibres, and fibre
460
reinforced composites has already been discussed in detail in Section 5.1.4.
461
462
Accepted for publication in Industrial Crops and Products on 13 July 2017.
18
5.3.2. Esterification
463
Esterification treatment is commonly used to reduce the hydrophilic character of natural fibres by forming ester
464
bonds between hydroxyl groups of cellulose rich fibres and carboxylic acid groups of other chemical reagents such as
465
acetic anhydride, maleic anhydride, and vinyl acetate. The esterification reaction may be conducted at room temperature
466
or at elevated temperature with or without catalysts (e.g. acids, pyridine or potassium carbonate) (Özmen 2012; Kabir et
467
al. 2013). Acetylation is one of the most commonly used esterification reactions to replace hydroxyl groups (OH) with
468
the acetyl groups (CH3CO) by using acetic anhydride (i.e. glacial acetic acid) or vinyl acetate as reagents.
469
Kabir et al. characterized the changes in chemical composition of hemp fibres after alkali (i.e. NaOH) pre-treatment
470
followed by acetyl treatment with glacial acetic acid at room temperature without catalysts (Kabir et al. 2013). Their
471
work confirmed that acetyl treatment with glacial acetic acid on hemp fibres can remove hemicellulose and lignin
472
constituents that cause increased thermal resistance (Kabir et al. 2013). Up to 6% ester content (g ester/ 100 g dry
473
matter of fibres) has been obtained after acetyl treatment of hemp fibres with acetic anhydride was performed for 120
474
min (Tserki et al. 2005). The removal of non-crystalline constituents of hemp fibres and slight decrease in fibre
475
crystallinity as a result of esterification were revealed using scanning electron microscopy (SEM) and X-ray diffraction
476
(XRD), respectively (Tserki et al. 2005).
477
Successful acetylation of hemp fibres with acetic anhydride in the presence of pyridine or potassium carbonate has
478
been achieved and this treatment resulted in a weight percentage gain (WPG) of 14% and 16%, respectively (Özmen
479
2012), whereas a WPG of 15% was achieved from acetyl treatment of hemp fibres with vinyl acetate in the presence of
480
potassium carbonate, but no weight gain (0% WPG) was obtained when pyridine was used as catalyst (Özmen 2012).
481
After hemp fibres were acetylated, the maximum weight loss temperature increased about 15 °C when compared with
482
untreated hemp fibres (Özmen 2012) .
483
The effect of surface treatment of hemp fibres using acetic anhydride and maleic anhydride on interfacial shear
484
strength (IFSS) of hemp fibre reinforced polylactide (PLA) and unsaturated polyester has also been investigated: The
485
IFFS of hemp fibre/PLA composites increased after acetic anhydride treatment. The increased IFFS from acetic
486
anhydride treatment was explained as being due to improved bonding between PLA and hemp fibres and increased PLA
487
transcrystallinity (Sawpan et al. 2011b). The IFSS of hemp fibre/unsaturated polyester also increased after both acetic
488
anhydride and maleic anhydride treatments, and the increased IFFS was interpreted to be due to improvement in
489
chemical bonding between the treated hemp fibres and the unsaturated polyester (Sawpan et al. 2011b).
490
Accepted for publication in Industrial Crops and Products on 13 July 2017.
19
5.3.3. Graft co-polymerization
491
Wettability of fibres by matrix polymers can be increased by graft copolymerizing short-chain molecules and
492
polymers onto the fibre surface using silane coupling agents or other copolymers or functional molecules such as
493
isocyanatoethyl methacrylate (IEM) (Chen et al. 2013), glycidyl methacrylate (GMA) (Moawia et al. 2016), methyl
494
methacrylate (MMA) (Singha and Rana 2010), or alkyl gallates (Ni et al. 2015).
495
Alkoxy silane is a multifunctional molecule which is commonly used as a coupling agent to modify natural fibre
496
surfaces (Alix et al. 2009; Sawpan et al. 2011a; Sawpan et al. 2011b; Kabir et al. 2013). Alkoxy silanes can form
497
chemical bonds with hydroxyl groups on fibre surfaces. During silane treatment, silanes undergo hydrolysis,
498
condensation and the bond formation stage. In the first step, silanols are formed after alkoxy silanes are hydrolyzed in
499
the presence of moisture (Sreekala et al. 2000). The silanols then react with the hydroxyl groups on fibre surface in the
500
condensation stage and eventually forms chemical bonds (i.e. fibre – O − Si).
501
Kabir et al. (Kabir et al. 2013) assessed the effect of silane treatment after alkali pre-treatment on chemical
502
composition and physical properties of hemp fibres and found that silane formed coverings on the fibre surfaces and
503
filled the spaces between the microfibrils without effects on chemical composition of hemp fibres (Kabir et al. 2013).
504
Sawpan et al. observed increased interfacial shear strength (IFSS)of hemp fibres reinforced polylactide (PLA) and
505
unsaturated polyester (UPE) composites after silane treatments (Sawpan et al. 2011b). The increased IFSS from silane
506
treatments was ascribed to improved compatibility between hemp fibres and matrix polymers. The highest IFSS of 20.3
507
MPa was obtained for the combined sodium hydroxide and silane treated hemp fibre/UPE composites (Sawpan et al.
508
2011b). Furthermore, slight decrease in average tensile strength of fibres from 577 MPa for untreated fibres to 554 MPa
509
for silane treated fibres was observed. The decrease in average tensile strength of fibres was explained by the decrease
510
in cellulose crystallinity after silane treatment (Sawpan et al. 2011a). However, the average stiffness of fibres increased
511
to 29.9 GPa after silane treatments, compared to 26.5 GPa for untreated fibres, and the increased stiffness was
512
presumably due to the removal of non-cellulosic components during silane treatments (Sawpan et al. 2011a).
513
After pre-treatment of hemp fibres with alkali, silane treatments have not led to apparent improvements in
514
mechanical and physical properties of hemp fibre reinforced composites (Kabir et al. 2013; Panaitescu et al. 2016). This
515
was probably because only amorphous hydroxyl groups can react with the functional groups of the coupling agents.
516
When hemp fibre is pre-treated with alkali, less hydroxyl groups are accessible to react with the functional groups of
517
silanes. Therefore other studies have been made on the effects of silane treatments on mechanical properties of hemp
518
fibres without alkali pre-treatment (Lu and Mysore Bhogaiah 2011; Panaitescu et al. 2016). These studies showed that
519
Accepted for publication in Industrial Crops and Products on 13 July 2017.
20
without alkali pre-treatment, treatment with 5% (w/w) triethoxyvinylsilane for 3 hours could increase the thermal
520
stability of hemp fibres, and temperature at 50% weight loss of hemp fibres increased to 358 °C after silane treatment
521
from 337 °C for untreated fibres (Lu and Mysore Bhogaiah 2011). A comparison has previously been conducted on the
522
effect of silane treatments with three different silanes, (i: γ−aminopropyltriethoxysilane (APS), ii:
523
γ−glycidoxypropyltrimethoxysilane (GPS), and iii: γ−methacryloxypropyltrimethoxy‒silane (MPS)) without alkali pre-
524
treatment on mechanical properties of hemp fibres/polypropylene (PP) composites (Panaitescu et al. 2016). The highest
525
increase in composite stiffness was achieved when fibres were treated with MPS, followed by APS and then GPS
526
(Panaitescu et al. 2016). It was found that the stiffness of PP increased by 67% when PP was reinforced with MPS
527
treated hemp fibres, and by only 30% when PP was reinforced with untreated hemp fibres (Panaitescu et al. 2016). The
528
improvement in the mechanical properties of hemp fibre reinforced composites after silane treatment was mainly due to
529
the increased wettability of hemp fibres by polymers after silane treatment.
530
Besides silanes, other functional molecules have been used in fibre surface modification for improving wettability of
531
hemp fibres by hydrophobic polymers. In previous studies, a combination of 1, 6- diisocyanatohexane (DIH) and 2-
532
hydroxyethyl acrylate (HEA) was used to modify hemp fibre surfaces, and significantly increased tensile strength,
533
flexural modulus of rupture (MOR), and flexural modulus of elasticity (MOE) have been achieved in hemp fibre
534
reinforced unsaturated polyester composites (Qiu et al. 2011). The highest tensile strength of 65 MPa was obtained in
535
composites made from 3% DIH combined with 2% HEA treated hemp fibres (Qiu et al. 2011). The highest MOR of
536
approx. 110 MPa and MOE of approx. 7.8 GPa were obtained in composites made from 2% DIH combined with 1.4%
537
HEA treated hemp fibres (Qiu et al. 2011). Besides using DIH and HEA, grafting N-methylol acrylamide (NMA) onto
538
fibre surfaces in the presence of sulfuric acid as a catalyst has been found to increase tensile strength, MOR and MOE
539
of hemp fibre reinforced unsaturated polyester composites (Qiu et al. 2012). Treatment of hemp fibres with 3% (w/w)
540
of NMA increased the tensile strength, flexural strength and flexural modulus of composites by 42.0%, 92.9%, and
541
158.6%, respectively, when compared with untreated hemp fibre reinforced composites (Qiu et al. 2012).
542
Other coupling agents such as isocyanatoethyl methacrylate (IEM) have also been tested in hemp fibre surface
543
modification (Chen et al. 2013). IEM is a heterofunctional monomer with a reactive isocyanate group and a vinyl
544
polymerizable double bond, and the two function groups can react independently with other amorphous hydroxyl
545
groups or vinyl monomers on the fibre surface respectively (Thomas 1983). Increased interfacial adhesion was observed
546
from SEM graphs of tensile-fractured surfaces of hemp unsaturated polyester composites, as evidenced by fewer pull-
547
out holes observed on the fractured surface of the composites with IEM treated hemp fibres (Chen et al. 2013). For the
548
Accepted for publication in Industrial Crops and Products on 13 July 2017.
21
mechanical properties of composites with hemp fibres treated at varied concentrations of IEM, the optimum treatment
549
was 3% (w/w) IEM for hemp fibres in terms of tensile strength (105 MPa), flexural strength (150 MPa) and flexural
550
modulus (10 GPa).
551
The grafting reactions discussed above have usually been conducted in the presence of chemical catalysts (e.g.
552
dibutyltin dilaurate for IEM treatment, and sulfuric acid for NMA treatment) or at acidic conditions (e.g. pH 4 – 5 for
553
silane treatment). Recently, grafting of non-polar polymers onto fibre surfaces using laccase catalyzed oxidation has
554
been demonstrated (Fan et al. 2012; Ni et al. 2015). Laccase catalysis can induce grafting of phenolic monomers onto
555
hemp fibre surfaces through phenoxy radical generation by catalytic oxidation of hydroxyl groups (Fan et al. 2012).
556
When hydroquinone was used as phenolic monomers during laccase treatment of hemp fibres, tensile strength of hemp
557
fibre reinforced polylactide acid (PLA) composites increased to 19 MPa from 13 MPa after laccase treatment (Fan et al.
558
2012). Enzyme catalyzed grafting can be done at low temperature and at pH values close to neutral which minimizes
559
the negative effect of natural fibres. It can be expected that laccase-catalyzed modification on natural fibre surfaces will
560
attract more attention in the future.
561
5.3.4. Physical treatment
562
Physical treatment of natural fibres using corona, UV, plasma or gamma radiation can change structural and surface
563
properties of the fibres and thereby influence the mechanical bonding between fibres and matrix polymers.
564
Corona treatment is a physical surface modification technique that uses low temperature corona discharge plasma to
565
oxidize fibre surfaces and impart changes in the surface properties. Corona treatment could result in a surface oxidation
566
and etching effect, leading to an improvement of the interfacial compatibility between fibres and matrix polymers
567
(Faruk et al. 2012). The corona plasma is generated by supplying a low frequency high voltage to electrodes that usually
568
have sharp tips (Ragoubi et al. 2010). Tensile strength of hemp fibre/PP composites increased to 38 MPa from 29 MPa
569
after hemp fibres were treated with corona discharge, and stiffness (refers toYoung’s modulus or elastic modulus)
570
increased to 1215 MPa from 1079 MPa (Ragoubi et al. 2010). After corona treatment, increased interfacial bonding
571
between hemp fibres and polypropylene was achieved, as evidenced by fewer unbroken hemp fibres pulled out of the
572
fracture surfaces observed (Ragoubi et al. 2010). In general, modifications on hemp fibres by corona treatment include
573
increased surface roughness caused by the formation of micro-pits and cavities, which probably explained the improved
574
interfacial bonding between fibres and polypropylene (Ragoubi et al. 2010).
575
Ultraviolet (UV) light is known for its shorter wavelength and lower energy than visible light. Eposing natural fibres
576
to UV radiation has been widely used in natural fibre surface modification (Mukhopadhyay and Fangueiro 2009; Olaru
577
Accepted for publication in Industrial Crops and Products on 13 July 2017.
22
et al. 2016). UV light (preferably at wavelength of 185 and 254 nm) in the presence of oxygen generates atomic oxygen
578
and ozone that results in the fibre surface being more hydrophilic (Zelez 1992). The behavior of hemp fibres when
579
subjected to a hydrothermal pre-treatment followed by exposure to UV radiation (200 nm < wavelength < 700 nm,
580
incident light intensity: 3.9 ×10-5 kW/cm2) for 120 – 500 hours was assessed. It was found that UV radiation can affect
581
the network of intra- and intermolecular hydrogen bonds and that there was a reduction of cellulose crystallinity caused
582
by UV radiation treatment (Olaru et al. 2016). UV radiation treatment on other types of natural fibres such as jute fibres
583
has also been carried out previously (Gassan and Gutowski 2000). In a study by Gassan et al. (Gassan and Gutowski
584
2000) tossa jute single fibres and fibre yarns were treated with UV radiation at a wavelength of 254 nm, which is the
585
most important wavelength regarding generating atomic oxygen and ozone (Zelez 1992). The UV radiation treatment
586
could lead to higher polarities (up to 200% increase) of the fibre surface, so that the wettability of the fibres by epoxy
587
matrix was improved (Gassan and Gutowski 2000). Up to 30% increase in composite flexural strength has been
588
achieved by UV treatment (Gassan and Gutowski 2000). In principle, increased polarity of fibres after UV treatment is
589
disadvantageous for composite use especially when other hydrophobic polymers are used, while increased polarity of
590
fibres probably benefits grafting co-polymerization and esterification and results in higher grafting efficiency and
591
esterification efficiency.
592
Plasma treatment, especially cold plasma treatment, is another physical treatment to modify fibre surface without
593
affecting the bulk properties of fibres, and is similar to corona treatment. Plasma treatment can be performed under
594
vacuum and under atmospheric conditions (Mukhopadhyay and Fangueiro 2009). The hydrophilic nature of hemp fibres
595
was found to be reduced by the etching effect of plasma, which was presumably due to migration of hydrophobic
596
compounds from the fibre bulk to the surface (Baltazar-Y-Jimenez and Bismarck 2007). Increased hydrophobicity of
597
the fibre surface is favorable for the mechanical performance of hemp fibre reinforced composites.
598
Gamma rays are packets of high-frequency electromagnetic energy and have been also used for natural fibre surface
599
modification. Olaru et al. studied the changes in structure of hemp fibres after hydrothermal pre-treatment followed by
600
gamma radiation, and found a 25 kGy dose gamma irradiation resulted in an increase of the crystallinity to 71.5% from
601
64% for control samples (Olaru et al. 2016). Interfacial bonding increased between pineapple leaf fibre/jute fibre and
602
bisphenol-A matrix polymers in the hybrid composites after pineapple leaf fibres/jute fibres were exposed to gamma
603
radiation with 5 kGy radiation dosage. The flexural strength of the hybrid composites increased from 33 MPa to 57
604
MPa and the impact strength of the composites increased from 2.5 KJ/m2 to 3.0 KJ/m2 (Raghavendra et al. 2015).
605
Accepted for publication in Industrial Crops and Products on 13 July 2017.
23
5.4. Moisture uptake minimization
606
The hydrophilic nature of cellulosic fibres not only makes them less compatible with commonly used polymers,
607
which are usually hydrophobic, but also makes them vulnerable to moisture and biological contamination. When natural
608
fibres are used as reinforcing agents in composites, polymers form a strong barrier against moisture and biological
609
activities and generally protect fibres from swelling and degradation. However, the matrix polymers can only slow the
610
moisture diffusion process. Sooner or later (e.g. up to several months), moisture will penetrate through the polymer
611
matrix and contact fibres and cause fibre swelling, leading to decrease in mechanical properties of the composites
612
(Dhakal et al. 2007; Christian and Billington 2012). A major restriction in the successful use of natural fibres in durable
613
polymer composite applications is that the moisture absorption causes physical property changes, especially fibre
614
dimensions change. In addition moisture absorption may introduce delamination and defects at interfaces, which make
615
natural fibres less favorable competitors of synthetic fibres for long-term performance and durability applications
616
(Baltazar-Y-Jimenez and Bismarck 2007). Therefore, fibre pre-treatments are expected to prolong the duration of NFCs
617
and stabilize their mechanical properties by reducing the moisture effect. Many fibre treatments have been attempted to
618
increase water resistance ability of NFCs. Some treatments have been shown to be efficient in minimizing natural fibre
619
moisture uptake, such as hydrothermal treatment (Rouison et al. 2005) and chemical modification of fibre surface (Alix
620
et al. 2009; Chen et al. 2013).
621
In natural fibres, pectic substances and hemicelluloses of fibres, along with other cell wall components, are the
622
primary components contributing to surface hydrophilicity (George et al. 2016) which leads to moisture absorption, as
623
indicated by the relative moisture absorption propensity shown in Table 8. Therefore the removal of the components
624
which are more inclined to absorb moisture can improve moisture resistance of natural fibres. George et al (George et
625
al. 2016) reported that the removal of pectin and hemicellulose was found to improve moisture resistance of resulting
626
hemp fibre reinforced polypropylene (PP) composites. All investigated enzymatic treatments using xylanase (Xyl),
627
polygalacturonase (PG), laccase (Lac) and a combination of xylanase and cellulase (Xyl + Cel) were shown to decrease
628
moisture uptake of the resulting fibre reinforced composites relative to untreated fibre reinforced composite. Xyl + Cel
629
and PG treatments provided the greatest reduction in moisture uptake (less than 0.6% after immersion in water for 4
630
weeks) for hemp fibre reinforced PP composites.
631
Bismarck et al. attempted to understand the moisture absorption behavior of hemp fibres by measuring ζ-potential of
632
fibres using 1 mM KCl after exposure to 100% humidity for varied durations, namely the time-dependence of the ζ-
633
potential (i.e. ζ-potential =f(t)) (Bismarck et al. 2002). The ζ-potential measurements could clearly differentiate the
634
Accepted for publication in Industrial Crops and Products on 13 July 2017.
24
degree of hydrophilicity of the natural fibres. The untreated hemp fibres are the most hydrophobic due to presence of
635
the highest amount of waxy substances on fibre surfaces, whereas the retting process decreased the water resistance
636
ability of hemp fibres through the removal of those hydrophobic coverings during retting (Bismarck et al. 2002). As
637
regards flax fibres, it was found the commercially available autoclave-treated (Duralin® process) fibres are more
638
hydrophobic than the original flax fibres. In this hydrothermal treatment, a primary depolymerization of hemicellulose
639
and lignin occurs after 30 min hydrothermal pre-treatment at 160 °C. Then, the aldehyde and phenolic functionalities
640
are chemically grafted onto fibre surfaces by the post-curing at 150 °C (Alix et al. 2014). Duralin® fibres were thus
641
found to have lower water uptake (approx. 18%) than untreated fibres (up to 43%) (Bismarck et al. 2002). Besides
642
Duralin® process, traditional autoclave treatment at mild conditions (i.e. 0, 0.5 and 1 bar) was found to increase water
643
resistance of natural fibres due to modification of fibre surfaces at low operating pressures, and at elevated operating
644
pressures internal fibre structures were altered through removal of pectins (Alix et al. 2014).
645
Fibre surface modification by grafting hydrophobic chemicals or short polymers has been a commonly used method
646
to reduce hydrophilic features of natural fibres. Qiu et al. grafted N-methylol acrylamide (NMA) in the presence of
647
sulfuric acid as catalyst onto hemp fibre surfaces. They found that the water uptake of the resulting fibres was
648
significantly decreased to 17% after fibres were treated with 10% (w/w) NMA compared to 22% for control samples. In
649
another study, Qiu et al. (Qiu et al. 2011) treated hemp fibres with a combination of 1,6-diisocyanatohexane (DIH) and
650
2-hydroxyethyl acrylate (HEA) and achieved increased water resistance of the resulting hemp/unsaturated polyester
651
(UPE) composites. For all investigated composite specimens, it took approx. 12 days for moisture diffusion to reach the
652
equilibrium, as indicated by the changes in water uptake of the composite specimens. However, composites made with
653
DIH-HEA treated fibres had significantly lower maximum water uptake (16 – 18%) than the composites made with
654
untreated fibres (approx. 22%), and water uptake tended to decrease with the increase in the concentration of DIH (0 ‒
655
5%, wt %) and HEA (0 – 3.5%, wt %) used for fibre treatments (Qiu et al. 2011).
656
5.5. Anti-microbial degradation
657
As discussed above, natural fibres are sensitive to biological degradation, which is another restriction on the
658
successful use of natural fibres in durable polymer composite applications. The degradation of cellulosic fibres by
659
microorganisms in nature is usually accompanied with moisture uptake, or results inevitably after natural fibres have
660
absorbed a high amount of moisture. The increase in moisture resistance of natural fibres in turn will also protect them
661
from biological degradation. In principal, the methods discussed in Section 5.4 can not only improve moisture resistant
662
of natural fibres, but also increase their ability to resist microbial degradation.
663
Accepted for publication in Industrial Crops and Products on 13 July 2017.
25
Pectin and hemicellulose are due to their moisture uptake propensity the most sensitive components to biological
664
degradation, as indicated by the biological degradation propensity shown in Table 8. In principle, the removal of those
665
components can lead to increased resistance to microbial degradation. Furthermore, such modifications of the fibre
666
surface have long been applied to wood and some plant fibres to improve dimensional stability and environmental
667
degradation resistance (Pott 2002). Hill et al. attempted to improve the resistance of NFCs to microbiological decay by
668
acetylation treatment of natural fibres (e.g. coir, flax and jute fibres) (Hill et al. 1998): Both unsterile soil and
669
vermiculite soil-layer tests were employed to determine the bio-resistance of chemically modified fibres (Hill et al.
670
1998). In microbiological decay tests, it was found that modified fibres exhibited a high degree of decay resistance in
671
both tests over a 5 month test period, whereas control samples failed in less than 1 month (Hill et al. 1998).
672
6. NFCs application
673
6.1. Durability of NFCs
674
Durability may be defined as "capable of lasting" or "not transitory". In the context of engineering materials,
675
durability would normally address resistance to degradation by chemicals (especially liquid water and moisture vapour),
676
biota or stress (especially creep and fatigue) or combinations of those factors (e.g. environmental fatigue). For metals
677
the resistance to chemicals is designated as resistance to corrosion, but for polymer composites the appropriate terms
678
would be degradation or deterioration. The durability of reinforced plastics has been reviewed by Pritchard (Pritchard
679
1998), Schutte (Schutte 1994) and Liao et al (Liao et al. 1998). Maxwell et al (Maxwell et al. 2005) have reviewed
680
accelerated ageing methods and lifetime prediction techniques for polymeric materials. For predicting suitability of a
681
material over the complete lifetime of components Highly Accelerated Life Testing (HALT) can be used. Degradation
682
mechanisms normally occur faster at higher temperatures, but if there is a change in the mechanism involved, then the
683
data obtained may be misleading. ASTM D5229-92 recommends that HALT should only be conducted up to
684
temperatures of 25°C below Tg (Standard 2014).
685
However, the above publications are primarily addressed to issues with synthetic fibre reinforced polymer matrix
686
composites. Summerscales (Summerscales 2014) has considered the issues surrounding bio-based composites in wet
687
conditions. Bajwa et al (Bajwa et al. 2015) have reviewed the impact of biofibres and coupling agents on the weathering
688
characteristics of composites. A particular issue for composites in the marine environment is osmosis and blistering
689
which leads to cosmetic and hydrodynamic damage, but to date no scientific literature on this topic has been identified.
690
At the macroscopic scale, the diffusion of gases, vapours or liquids into a material is normally modelled using Fick’s
691
law, with the fluid (e.g. moisture) content initially increasing with exposure time then approaching a saturation level.
692
Accepted for publication in Industrial Crops and Products on 13 July 2017.
26
For systems where sorption and/or reaction–diffusion produce a non-Fickian response, the diffusion coefficient can be
693
directly derived from sorption isotherms such as Henry’s or Langmuir’s laws (Verdu and Colin 2012).
694
Derrien and Gilormini (DG) (Derrien and Gilormini 2009) have presented a model for the time-dependent evolution
695
of the moisture content during unidirectional diffusion in a polymer submitted to hydrostatic load. Jacquemin and
696
Fréour (JF) (Jacquemin and Fréour 2012) presented two multiphysics models (a thermodynamic approach and a free
697
volume theory) for the effects of plasticization during water sorption and the internal mechanical state profile at both
698
constituent and ply scales. Discrepancies between the DG and JF models were found to increase significantly with the
699
coefficient of moisture expansion (CME). Perreux (Perreux 2012) presented a general behavior model to account for
700
time-dependent mechanical and environmental (water) loading of laminates and the induced damage. The first-stage
701
micro-mesomodel describes the variation in stiffness due to microcracks. The second stage kinetic model is based on
702
the thermodynamic definition of the forces driving damage and other dissipation potentials.
703
Osmosis is the process by which solution strengths are equalized by passage of the solvent (usually water) through a
704
semi-permeable membrane. In fibre-reinforced composites, the polymer matrix can act as the membrane. As water
705
diffuses through the polymer, any soluble solid material can dissolve and thus form a strong solution. Water will then
706
diffuse to this solution until the concentration gradient is reduced to zero. The volume of the solution will increase with
707
dilution and exert pressure on the surrounding material. When the stresses exceed a critical level, delamination will
708
occur (normally at the gelcoat-laminate interface) and will be manifest as blisters on the surface of the laminate. A
709
comprehensive list of chemical factors implicated in osmosis and blistering and other measures to reduce or eliminate
710
the problem are given in (Searle and Summerscales 1999).
711
A potential problem with natural fibre-reinforced polymer matrix composites is the hydrophilic nature of the ligno-
712
cellulose fibres and hence the moisture sensitivity of the resulting composites. Embedding the hydrophilic fibres in a
713
hydrophobic matrix will delay the absorption of water but diffusion and damage may compromise the material over
714
extended periods of time. Moisture will induce dimensional changes (swelling), mechanical performance changes
715
(plasticization and hence higher strains to failure but lower moduli) and higher susceptibility to microbiological attack.
716
Costa and D’Almeida (Costa and D’Almeida 1999) studied the effect of water absorption on the flexural properties
717
of jute or sisal fibre reinforced polyester or epoxy matrix composites. The diffusion behavior in both composites could
718
be described by the Fickian model. The jute-epoxy composites showed the best mechanical properties for all immersion
719
times studied. This behavior was attributed to better fibre–matrix interface integrity with epoxy resin and better
720
moisture resistance of the jute fibres (Costa and D’Almeida 1999).
721
Accepted for publication in Industrial Crops and Products on 13 July 2017.
27
Shah et al. (Shah et al. 2012; Shah et al. 2013) have recently presented fatigue life evaluations for aligned plant
722
fibre composites through S–N curves and constant-life diagrams. The normalized fatigue performance (the fatigue
723
strength exponent, b, defined by the stress intercept at twice the number of load reversals to failure) of the natural fibre
724
composites was found to lie between that of glass-fibre and carbon-fibre reinforced composites.
725
There is increasing interest in the use of bio-based and/or biodegradable thermoplastic polymers or thermosetting
726
resins as the matrix for composites with varying proportions of precursor materials extracted from plants (see e.g.
727
(Summerscales and Grove 2013)). Ishimaru et al. (Ishimaru et al. 2012) reported that polycaprolactone,
728
polyhydroxyalkanoate and polylactide thermoplastics or their copolymers (amongst others) have been applied as self-
729
polishing/exfoliating matrices for controlled release of antifouling compositions. They conducted experiments which
730
showed that barnacle settlement was significantly reduced by the slow release of low molecular weight poly (L-lactic
731
acid) (PLLA without antifoulant chemical) in natural seawater.
732
6.2. Application of NFCs
733
NFCs are used in various applications nowadays. They are used in internal components in the automotive industry,
734
e.g. kenaf/PLA spare tyre cover for Toyota Raum (2003) (Zini and Scandola 2011), dashboards, armrests and seat
735
components (Daimler Chrysler) (Shuda et al. 2008), door linings and panels (BMW) (Holbery and Houston 2006;
736
Shuda et al. 2008), door panel inserts and package trays (Opel) (Shuda et al. 2008), door panels (Ford) (Shuda et al.
737
2008) and floor trays (Faruk et al. 2014), pillar panels and central consoles (Volvo) (Faruk et al. 2014). Other
738
automotive applications include boot (US trunk) liners, door cassettes, parcel shelves, and head restraints (Faruk et al.
739
2014). The high damping characteristics of NFCs will be appropriate for noise, vibration and harshness (NVH)
740
suppression in all transport vehicles, such as noise insulation panels (Faruk et al. 2014). Applications on exterior panels
741
of vehicles may be compromised by the moisture sensitivity of the respective composite components. It has been shown
742
that natural fibres are a potential replacement for synthetic fibres in composite brake pads (Chand and Fahim 2008). In
743
the construction industry, natural fibre biocomposites are being considered for hand rails, parquet flooring, slabs and
744
window frames (Staiger and Tucker 2008). Amongst electronic applications, NEC developed a kenaf/PLA mobile
745
phone outer casing in 2006 (Staiger and Tucker 2008).
746
Although there are challenges in making composites suitable for the marine environment, especially for vessels
747
which remain afloat at all times, there have been a number of demonstrator applications. In the aerospace industry, the
748
most likely first applications would be to exploit the high specific stiffness of composites in in-cabin components such
749
as seats and luggage lockers, provided the flame, smoke and toxicity (FST) requirements can be met. Other high-value
750
Accepted for publication in Industrial Crops and Products on 13 July 2017.
28
applications might include biomedical and funeral articles, furniture, instrument cases and packaging (Zini and
751
Scandola 2011). In the future, NFCs will see significantly increased use in structural applications when issues that limit
752
successful use of natural fibres in composites have been solved.
753
7. Perspectives of natural fibre pre-treatment for composite use
754
Most fibre pre-treatment methods discussed above are chemical reaction-based methods, while more use of enzymes,
755
especially oxidases e.g. laccases, for fibre pre-treatment is emerging. Those enzymatic fibre pre-treatment methods are
756
environmentally friendly and more targeted for fibre treatment or surface modification. However, the mechanisms
757
behind enzymatic reactions involving solid substrates (i.e. plant fibres) during fibre treatments are not fully understood.
758
Therefore further work is still needed to improve understanding of how enzymes modify plant fibres to optimize those
759
enzyme treatments and to develop more targeted methods.
760
Furthermore, the most widely used plant fibres for NFCs are bast fibres of plant such as hemp and flax because these
761
bast fibres are strong and long. The long fibres could be easily aligned to make fibre orientation more uniform during
762
composite processing. However, the cellulosic rich fibres from other lignocellulosic materials (e.g. wheat bran and
763
wheat straw) are rather short and the short fibres limit their applications in producing strong NFCs. Therefore a
764
continuous form of reinforcement agents (sliver, yarn, filaments) is needed from various sources of lignocellulosic
765
materials for producing strong biocomposites in industrial use.
766
Studies are emerging on how to produce continuous form of filaments (also called regenerated fibres) using ionic
767
liquids as dissolution solvent (Ma et al. 2015; Michud et al. 2015). Ionic liquids (ILs) have some unique, advantageous
768
properties such as low toxicity, low evaporation and high thermal stability. More importantly, ILs have a high cellulose
769
solubility at moderate conditions. This characteristic provides a useful tool to reshape and modify the components and
770
properties of the resulting filaments. It has been shown that regenerated filaments with high tensile strength up to 800
771
MPa can be achieved with pure cellulose under optimum operating conditions using dry-jet-wet spinning (Ma et al.
772
2015). However, the surface properties of these regenerated fibres have not been intensively studied and little
773
information is available currently. If the regenerated fibres need to be pre-treated prior to being used as reinforcing
774
agents in composites, most of above discussed surface modification methods can in principle be applied to these fibres
775
because the chemical composition of those natural plant fibres and regenerated plant fibres are more or less similar.
776
Accepted for publication in Industrial Crops and Products on 13 July 2017.
29
8. Conclusions
777
Many attempts have been made to use natural plant fibres in polymer reinforced composite materials. Improved
778
mechanical performance and durability of NFCs have been achieved via improved fibre cultivation, fibre selection,
779
fibre extraction, fibre treatments, and fibre surface modification methods. This paper has reviewed the research that has
780
focused on improved short and long term mechanical properties and durability of NFCs containing natural fibres,
781
notably differently pre-treated hemp fibres, focusing on improved interfacial bonding between fibres and matrix
782
polymers and decreased hydrophilic features of fibres resulting from different fibre pre-treatments. Currently, many
783
novel fibre pretreatment strategies and surface modification methods are chemically based methods, while research on
784
use of enzymes, especially pectinases and oxidases e.g. laccases, for plant fibre pre-treatment and surface modification
785
are emerging. Future research is still needed to expedite the application of NFCs and extend their durability and to
786
improve their long-term performance including improvement of moisture resistance and microbial resistance by
787
development of new targeted and environmentally friendly fibre pre-treatment methods.
788
789
9. Acknowledgement
790
The authors thank the Danish Council for Independent Research supporting the CelFiMat project (No. 0602-02409B:
791
“High quality cellulosic fibres for strong biocomposite materials”) and the Dept. of Chemical and Biochemical
792
Engineering, Technical University of Denmark, for financially supporting this work.
793
794
Accepted for publication in Industrial Crops and Products on 13 July 2017.
30
References
795
Adamsen APS, Akin DE, Rigsby LL (2002) Chelating agents and enzyme retting of flax. Text Res J 72:296–302.
796
Alix S, Colasse L, Morvan C, Lebrun L, Marais S (2014) Pressure impact of autoclave treatment on water sorption and
797
pectin composition of flax cellulosic-fibres. Carbohydr Polym 102:21–29.
798
Alix S, Philippe E, Bessadok A, Lebrun L, Morvan C, Marais S (2009) Effect of chemical treatments on water sorption
799
and mechanical properties of flax fibres. Bioresour Technol 100:4742–4749.
800
Álvarez C, Rojano B, Almaza O, Rojas OJ, Gañán P (2011) Self-bonding boards from plantain fibre bundles after
801
enzymatic treatment: adhesion improvement of lignocellulosic products by enzymatic pre-treatment. J Polym
802
Environ 19:182–188.
803
Amaducci S, Scordia D, Liu FH, Zhang Q, Guo H, Testa G, Cosentino SL (2014) Key cultivation techniques for hemp
804
in Europe and China. Ind Crops Prod 68:2–16.
805
Aziz SH, Ansell MP (2004) The effect of alkalization and fibre alignment on the mechanical and thermal properties of
806
kenaf and hemp bast fibre composites: Part 1 - polyester resin matrix. Compos Sci Technol 64:1219–1230.
807
Bacci L, Di Lonardo S, Albanese L, Mastromei G, Perito B (2010) Effect of different extraction methods on fibre
808
quality of nettle (Urtica dioica L.). Text Res J 81:827–837.
809
Bajwa DS, Bajwa SG, Holt GA (2015) Impact of biofibres and coupling agents on the weathering characteristics of
810
composites. Polym Degrad Stab 120:212–219. doi: 10.1016/j.polymdegradstab.2015.06.015
811
Bakare IO, Okieimen FE, Pavithran C, Abdul Khalil HPS, Brahmakumar M (2010) Mechanical and thermal properties
812
of sisal fibre-reinforced rubber seed oil-based polyurethane composites. Mater Des 31:4274–4280.
813
Baley C (2002) Analysis of the flax fibres tensile behaviour and analysis of the tensile stiffness increase. Compos Part
814
A Appl Sci Manuf 33:939–948.
815
Baley C, Busnel F, Grohens Y, Sire O (2006) Influence of chemical treatments on surface properties and adhesion of
816
flax fibre-polyester resin. Compos Part A Appl Sci Manuf 37:1626–1637.
817
Baltazar-Y-Jimenez A, Bismarck A (2007) Surface modification of lignocellulosic fibres in atmospheric air pressure
818
plasma. Green Chem 9:1057–1066. doi: doi: 10.1039/B618398K
819
Behazin E, Ogunsona E, Rodriguez-uribe A, Mohanty AK, Misra M, Anyia AO (2016) Mechanical, chemical, and
820
physical properties of wood and perennial grass biochars for possible composite application. Bioresources
821
11:1334–1348.
822
Benen JAE, Kester HCM, Visser J (1999) Kinetic characterization of Aspergillus niger N400 endopolygalacturonases I,
823
Accepted for publication in Industrial Crops and Products on 13 July 2017.
31
II and C. Eur J Biochem 259:577–585.
824
Benko Z, Siika-aho M, Viikari L, Réczey K (2008) Evaluation of the role of xyloglucanase in the enzymatic hydrolysis
825
of lignocellulosic substrates. Enzyme Microb Technol 43:109–114.
826
Bessadok A, Marais S, Gouanvé F, Colasse L, Zimmerlin I, Roudesli S, Métayer M (2007) Effect of chemical
827
treatments of Alfa (Stipa tenacissima) fibres on water-sorption properties. Compos Sci Technol 67:685–697.
828
Bismarck A, Aranbefwi-Askargorta I, Springer J, Lampke T, Wielage B, Stamboulis A, Shenderovich I, Limbach H-H
829
(2002) Surface characterization of flax, hemp and cellulose fibres; surface properties and the water uptake
830
behavior. Polym Compos 23:872–894.
831
Bliznakov ED, White CC, Shaw MT (2000) Mechanical properties of blends of HDPE and recycled urea-formaldehyde
832
resin. J Appl Polym Sci 77:3220–3227.
833
Bonatti PM, Ferrari C, Focher B, Grippo C, Torri G, Cosentino C (2004) Histochemical and supramolecular studies in
834
determining quality of hemp fibres for textile applications. Euphytica 140:55–64.
835
Bourmaud A, Le Duigou A, Gourier C, Baley C (2016) Influence of processing temperature on mechanical performance
836
of unidirectional polyamide 11 - flax fibre composites.
837
Bourmaud A, Morvan C, Bouali A, Placet V, Perré P, Baley C (2013) Relationships between micro-fibrillar angle,
838
mechanical properties and biochemical composition of flax fibres. Ind Crops Prod 44:343–351.
839
Brahim S Ben, Cheikh R Ben (2007) Influence of fibre orientation and volume fraction on the tensile properties of
840
unidirectional Alfa-polyester composite. Compos Sci Technol 67:140–147.
841
Callister WD (1994) Materials science and engineering, 3rd edn. John Wiley & Sons, INC., New York
842
Carpita NC, Gibeaut DM (1993) Structural models of primary cell walls in flowering plants: consistency of molecular
843
structure with the physical properties of the walls during growth. Plant J 3:1–30.
844
Chand N, Fahim M (2008) Tribology of natural fibre polymer composites. Woodhead Publishing Limited, Cambridge
845
Charlet K, Baley C, Morvan C, Jernot JP, Gomina M, Bréard J (2007) Characteristics of Hermès flax fibres as a
846
function of their location in the stem and properties of the derived unidirectional composites. Compos Part A
847
Appl Sci Manuf 38:1912–1921.
848
Chen TT, Liu WD, Qiu RH (2013) Mechanical properties and water absorption of hemp fibres-reinforced unsaturated
849
polyester composites: effect of fibre surface treatment with a heterofunctional monomer. BioResources 8:2780–
850
2791.
851
Chesson A, Gordon AH, Lomax JA (1983) Substituent groups linked by alkali-labile bonds to arabinose and xylose
852
Accepted for publication in Industrial Crops and Products on 13 July 2017.
32
residues of legume, grass and cereal straw cell walls and their fate during digestion by rumen microorganisms. J
853
Sci Food Agric 34:1330–1340.
854
Christian SJ, Billington SL (2012) Moisture diffusion and its impact on uniaxial tensile response of biobased
855
composites. Compos Part B Eng 43:2303–2312. doi: 10.1016/j.compositesb.2011.11.063
856
Cierpucha W, Kozłowski R, Mańkowski J, Waśko J, Mańkowski T (2004) Applicability of flax and hemp as raw
857
materials for production of cotton-like fibres and blended yarns in Poland. Fibres Text East Eur 12:13–18.
858
Costa FHMM, D’Almeida JRM (1999) Effect of Water Absorption on the Mechanical Properties of Sisal and Jute Fibre
859
Composites. Polym Plast Technol Eng 38:1081–1094. doi: 10.1080/03602559909351632
860
Crônier D, Monties B, Chabbert B, Cronier, D ; Monties, B; Chabbert B (2005) Structure and chemical composition of
861
bast fibres isolated from developing hemp stem. J Agric Food Chem 53:8279–8289.
862
Crosky A, Soatthiyanon N, Ruys D, Meatherall S, Potter S (2014) Thermoset matrix natural fibre-reinforced
863
composites. In: Hodzic A, Shanks R (eds) Natural Fibre Composites. Woodhead Publishing Limited, pp 233–270
864
Da Porto C, Decorti D, Natolino A (2015) Potential oil yield, fatty acid composition, and oxidation stability of the
865
hempseed oil from four Cannabis sativa L. cultivars. J Diet Suppl 12:1–10.
866
Dai D (2010) Characteristic and performance of elementary hemp fibre. Mater Sci Appl 1:336–342.
867
Deanin RD, Mead JL (2007) Synthetic resins and plastics. In: Kent JA (ed) Kent and Riegel’s Handbook of Industrial
868
Chemistry and Biotechnology, 11th ed. springer, Germany, pp 623–687
869
Derrien K, Gilormini P (2009) The effect of moisture-induced swelling on the absorption capacity of transversely
870
isotropic elastic polymer-matrix composites. Int J Solids Struct 46:1547–1553. doi: 10.1016/j.ijsolstr.2008.11.014
871
Dhakal HN, Zhang ZY, Richardson MOW (2007) Effect of water absorption on the mechanical properties of hemp fibre
872
reinforced unsaturated polyester composites. Compos Sci Technol 67:1674–1683.
873
Di Candilo M, Bonatti PM, Guidetti C, Focher B, Grippo C, Tamburini E, Mastromei G (2010) Effects of selected
874
pectinolytic bacterial strains on water-retting of hemp and fibre properties. J Appl Microbiol 108:194–203.
875
Domínguez JC, Madsen B (2015) Development of new biomass-based furan / glass composites manufactured by the
876
double-vacuum-bag technique. J Compos Mater 49:2993–3003.
877
Duc F, Bourban PE, Plummer CJG, Månson JAE (2014) Damping of thermoset and thermoplastic flax fibre
878
composites. Compos Part A Appl Sci Manuf 64:115–123.
879
Duval A, Bourmaud A, Augier L, Baley C (2011) Influence of the sampling area of the stem on the mechanical
880
properties of hemp fibres. Mater Lett 65:797–800.
881
Accepted for publication in Industrial Crops and Products on 13 July 2017.
33
Džalto J, Medina L, Mitschang P (2014) Volumetric interaction and material characterization of flax/furan bio-
882
composites. KMUTNB Int J Appl Sci Technol 7:11–21.
883
Eisenreich N (2008) New classes of Engineering Composites Materials from Renewable Resources.
884
Evans JD, Akin DE, Foulk JA (2002) Flax-retting by polygalaxturonase-containing enzyme mixtures and effects on
885
fibre properties. J Biotechnol 97:223–231.
886
Fan M, Dai D, Yang A (2011) High strength natural fibre composite: defibrillation and its mechanisms of nanocellulose
887
hemp fibres. Int J Polym Mater Polym Biomater 60:1026–1040.
888
Fan X, Wang Q, Zhou C, Wang P, Yuan J, Cui L (2012) Laccase-induced hemp fibre grafted phenolic monomer used to
889
enhance the performance of resin composite. 1–6.
890
Faruk O, Bledzki AK, Fink H-P, Sain M (2014) Progress Report on Natural Fibre Reinforced Composites. Macromol
891
Mater Eng 299:9–26. doi: 10.1002/mame.201300008
892
Faruk O, Bledzki AK, Fink HP, Sain M (2012) Biocomposites reinforced with natural fibres: 2000-2010. Prog Polym
893
Sci 37:1552–1596.
894
Fu SY, Lauke B, Mäder E, Yue CY, Hu X (2000) Tensile properties of short-glass-fibre- and short-carbon-fibre-
895
reinforced polypropylene composites. Compos Part A Appl Sci Manuf 31:1117–1125.
896
Fuqua MA, Huo S, Ulven CA (2012) Natural Fibre Reinforced Composites. Polym Rev 52:259–320.
897
Gassan J, Gutowski VS (2000) Effects of corona discharge and UV treatment on the properties of jute-fibre expoxy
898
composites. Compos Sci Technol 60:2857–2863. doi: 10.1016/S0266-3538(00)00168-8
899
George M, Mussone PG, Alemaskin K, Chae M, Wolodko J, Bressler DC (2016) Enzymatically treated natural fibres as
900
reinforcing agents for biocomposite material: mechanical, thermal, and moisture absorption characterization. J
901
Mater Sci 51:2677–2686. doi: 10.1007/s10853-015-9582-z
902
Griko Y V (1999) Energetics of Ca(2+)-EDTA interactions: calorimetric study. Biophys Chem 79:117–127.
903
Hagstrand PO, Oksman K (2001) Mechanical properties and morphology of flax fibre reinforced melamine-
904
formaldehyde composites. Polym Compos 22:568–578.
905
Hauptt RA, Sellers T (1994) Characterizations of phenol-formaldehyde resol resins. Ind Eng Chem Res 33:693–697.
906
Herer J (1985) Hemp and the marijuana conspiracy. In: Cabarga L, Herer J, Duby RA (eds) The emperor wears no
907
clothes, 11th Eds. Ah Ha Publishing, California,
908
Hill CAS, Abdul Khalil HPS, Hale MD (1998) A study of the potential of acetylation to improve the properties of plant
909
fibres. Ind Crops Prod 8:53–63.
910
Accepted for publication in Industrial Crops and Products on 13 July 2017.
34
Holbery J, Houston D (2006) Natural-fibre-reinforced polymer composites in automotive applications. J Miner Met
911
Mater Soc 58:80–86.
912
Huang Z, Wang N, Zhang Y, Hu H, Luo Y (2012) Effect of mechanical activation pretreatment on the properties of
913
sugarcane bagasse/poly(vinyl chloride) composites. Compos Part A Appl Sci Manuf 43:114–120.
914
Hufenbach W, Gude M, Geller S, Czulak A (2013) Manufacture of natural fibre-reinforced polyurethane composites
915
using the long fibre injection process. Polimery/Polymers 58:473–475.
916
Ikeda R, Uyama H, Kobayashi S (1996) Novel synthetic pathway to a poly (phenylene oxide). laccase-catalyzed
917
oxidative polymerization of syringic acid. Macromolecules 29:3053–3054.
918
Ishimaru N, Tsukegi T, Wakisaka M, Shirai Y, Nishida H (2012) Effects of poly(L-lactic acid) hydrolysis on
919
attachment of barnacle cypris larvae. Polym Degrad Stab 97:2170–2176. doi:
920
10.1016/j.polymdegradstab.2012.08.012
921
Islam MS, Pickering KL, Foreman NJ (2011) Influence of alkali fibre treatment and fibre processing on the mechanical
922
properties of hemp/epoxy composites. J Appl Polym Sci 119:3696–3707.
923
Jacquemin F, Fréour S (2012) Water-mechanical property coupling. In: Ifremer-ONR Workshop on the durability of
924
composites in a matrine environment. pp 41–46
925
Jacquet N, Maniet G, Vanderghem C, Delvigne F, Richel A (2015) Application of Steam Explosion as Pretreatment on
926
Lignocellulosic Material: A Review. Ind Eng Chem Res 54:2593–2598. doi: 10.1021/ie503151g
927
Jankauskiene Z, Butkute B, Gruzdeviene E, Cesevičiene J, Fernando AL, Jankauskienė Z, Butkutė B, Gruzdevienė E,
928
Cesevičienė J, Fernando AL, Jankauskiene Z, Butkute B, Gruzdeviene E, Cesevičiene J, Fernando AL (2015)
929
Chemical composition and physical properties of dew- and water-retted hemp fibres. Ind Crops Prod 75:206–211.
930
Jankauskienė Z, Gruzdevienė E (2013) Physical parameters of dew retted and water retted hemp (Cannabis sativa L.)
931
fibres. Zemdirbyste-Agriculture 100:71–80.
932
John J, Mani SA, Glasg MR, Palaniswamy K, Ramanathan A, Aziz A, Razak A (2015) Flexural properties of
933
poly(methyl methacrylate) resin reinforced with oil palm empty fruit bunch fibres: a preliminary finding. J
934
Prosthodont 24:233–238.
935
Joshi S V., Drzal LT, Mohanty AK, Arora S (2004) Are natural fibre composites environmentally superior to glass fibre
936
reinforced composites? Compos Part A Appl Sci Manuf 35:371–376.
937
Kabir MM, Wang H, Lau KT, Cardona F (2013) Applied surface science effects of chemical treatments on hemp fibre
938
structure. Appl Surf Sci 276:13–23.
939
Accepted for publication in Industrial Crops and Products on 13 July 2017.
35
Keller A (2003) Compounding and mechanical properties of biodegradable hemp fibre composites. Compos Sci
940
Technol 63:1307–1316. doi: 10.1016/S0266-3538(03)00102-7
941
Keller A, Leupin M, Mediavilla V, Wintermantel E (2001) Influence of the growth stage of industrial hemp on chemical
942
and physical properties of the fibres. Ind Crops Prod 13:35–48.
943
Kiziltas EE, Yang HS, Kiziltas A, Boran S, Ozen E, Gardner DJ (2016) Thermal analysis of polyamide 6 composites
944
filled by natural fibre blend. BioResources 11:4758–4769. doi: 10.15376/biores.11.2.4758-4769
945
Komuraiah A, Kumar NS, Prasad BD (2014) Chemical composition of natural fibres and its influence on their
946
mechanical properties. Mech Compos Mater 50:359–376.
947
Korte S, Staiger MP (2008) Effect of processing route on the composition and properties of hemp fibre. Fibres Polym
948
9:593–603.
949
Kostic M, Pejic B, Skundric P (2008) Quality of chemically modified hemp fibres. Bioresour Technol 99:94–99.
950
Kukle S, Gravitis J, Putnina A, Stikute A (2011) The effect of steam explosion treatment on technical hemp fibres. Proc
951
8th Int Sci Pract Conf 1:230–237.
952
Le Troedec M, Sedan D, Peyratout C, Bonnet JP, Smith A, Guinebretiere R, Gloaguen V, Krausz P (2008) Influence of
953
various chemical treatments on the composition and structure of hemp fibres. Compos Part A Appl Sci Manuf
954
39:514–522.
955
Lefeuvre A, Bourmaud A, Morvan C, Baley C (2014) Elementary flax fibre tensile properties: Correlation between
956
stress-strain behaviour and fibre composition. Ind Crops Prod 52:762–769.
957
Lerche H, Benthien JT, Schwarz KU, Ohlmeyer M (2014) Effects of defibration conditions on mechanical and physical
958
properties of wood fibre/high-density polyethylene composites. J Wood Chem Technol 34:98–110.
959
Li Y, Pickering KL (2008) Hemp fibre reinforced composites using chelator and enzyme treatments. Compos Sci
960
Technol 68:3293–3298.
961
Li Y, Pickering KL, Farrell RL (2009) Determination of interfacial shear strength of white rot fungi treated hemp fibre
962
reinforced polypropylene. Compos Sci Technol 69:1165–1171.
963
Li A, Zhu Y, Xu L, Zhu W, Tian X (2008) Comparative study on the determination of assay for laccase of Trametes sp.
964
African J Biochem Res 2:181–183.
965
Liao K, Schultheisz CR, Hunston DL, Catherine Brinson L (1998) Long-term durability of fibre-reinforced polymer-
966
matrix composite materials for infrastructure applications: A review. J Adv Mater 30:3–40.
967
Liners F, Letesson JJ, Didembourg C, Van Cutsem P (1989) Monoclonal antibodies against pectin: recognition of a
968
Accepted for publication in Industrial Crops and Products on 13 July 2017.
36
conformation induced by calcium. Plant Physiol 91:1419–1424.
969
Liu M, Baum A, Odermatt J, Berger J, Yu L, Zeuner B, Thygesen A, Holck J, Meyer AS (2016a) Oxidation of lignin in
970
hemp fibres by laccase: effects on mechanical properties of hemp fibres and unidirectional fibre/epoxy
971
composites. Compos Part A Appl Sci In press DOI: 10.1016/j.compositesa.2017.01.026.
972
Liu M, Fernando D, Daniel G, Madsen B, Meyer A, Ale M, Thygesen A (2015a) Effect of harvest time and field retting
973
duration on the chemical composition, morphology and mechanical properties of hemp fibres. Ind Crops Prod
974
69:29–39.
975
Liu M, Fernando D, Meyer AS, Madsen B, Daniel G, Thygesen A (2015b) Characterization and biological
976
depectinization of hemp fibres originating from different stem sections. Ind Crops Prod 76:880–891.
977
Liu M, Meyer AS, Fernando D, Silva DAS, Geoffrey D, Thygesen A (2016b) Effect of pectin and hemicellulose
978
removal from hemp fibres on the mechanical properties of unidirectional hemp/epoxy composites. Compos Part A
979
Appl Sci Manuf 90:724–735.
980
Liu M, Silva DAS, Fernando D, Meyer AS, Madsen BB, Daniel G, Thygesen A (2016c) Controlled retting of hemp
981
fibres: Effect of hydrothermal pre-treatment and enzymatic retting on the mechanical properties of unidirectional
982
hemp/epoxy composites. Compos Part A Appl Sci Manuf 88:253–262. doi: 10.1016/j.compositesa.2016.06.003
983
Liu M, Ale MT, Kołaczkowski B, Fernando D, Daniel G, Thygesen A (2016d) Comparison of traditional field retting
984
and Phlebia radiata Cel 26 retting of hemp fibres for fibre-reinforced composites. (submitted manuscript).
985
Lu N, Mysore Bhogaiah S (2011) Effect of alkali and silane treatment on the thermal stability of hemp fibres as
986
reinforcement in composite structures. Appl Mech Mater 71–78:616–620. doi:
987
10.4028/www.scientific.net/AMM.71-78.616
988
Lu N, Oza S (2013) A comparative study of the mechanical properties of hemp fibre with virgin and recycled high
989
density polyethylene matrix. Compos Part B Eng 45:1651–1656.
990
Lupoi JS, Singh S, Parthasarathi R, Simmons BA, Henry RJ (2015) Recent innovations in analytical methods for the
991
qualitative and quantitative assessment of lignin. Renew Sustain Energy Rev 49:871–906.
992
Ma Y, Asaadi S, Johansson L-S, Ahvenainen P, Reza M, Alekhina M, Rautkari L, Michud A, Hauru L, Hummel M,
993
Sixta H (2015) High-strength composite fibres from cellulose-lignin blends regenerated from ionic liquid
994
solution. ChemSusChem 8:4030–4039.
995
Madsen B, Hoffmeyer P, Lilholt H (2007a) Hemp yarn reinforced composites - II. Tensile properties. Compos Part A
996
Appl Sci Manuf 38:2204–2215. doi: 10.1016/j.compositesa.2007.06.002
997
Accepted for publication in Industrial Crops and Products on 13 July 2017.
37
Madsen B, Hoffmeyer P, Thomsen AB, Lilholt H (2007b) Hemp yarn reinforced composites - I. Yarn characteristics.
998
Compos Part A Appl Sci Manuf 38:2194–2203.
999
Marais S, Gouanvé F, Bonnesoeur A, Grenet J, Poncin-Epaillard F, Morvan C, Métayer M (2005) Unsaturated polyester
1000
composites reinforced with flax fibres: effect of cold plasma and autoclave treatments on mechanical and
1001
permeation properties. Compos Part A Appl Sci Manuf 36:975–986.
1002
Markwalder HU, Neukom H (1976) Diferulic acid as a possible crosslink in hemicelluloses from wheat germ.
1003
Phytochemistry 15:836–837.
1004
Marrot L, Lefeuvre A, Pontoire B, Bourmaud A, Baley C (2013) Analysis of the hemp fibre mechanical properties and
1005
their scattering (Fedora 17). Ind Crops Prod 51:317–327.
1006
Mattinen ML, Suortti T, Gosselink R, Argyropoulos DS, Evtuguin D, Suurnäkki A, De Jong E, Tamminen T (2008)
1007
Polymerization of different lignins by laccase. BioResources 3:549–565.
1008
Maxwell AS, Broughton WR, Dean G, Sims GD (2005) Review of accelerated ageing methods and lifetime prediction
1009
techniques for polymeric materials.
1010
Mehmood S, Madsen B (2012) Properties and performance of flax yarn/thermoplastic polyester composites. J Reinf
1011
Plast Compos 31:1746–1757. doi: 10.1177/0731684412441686
1012
Michalak M, Thomassen L V., Roytio H, Ouwehand AC, Meyer AS, Mikkelsen JD (2012) Expression and
1013
characterization of an endo-1,4-β-galactanase from Emericella nidulans in Pichia pastoris for enzymatic design
1014
of potentially prebiotic oligosaccharides from potato galactans. Enzyme Microb Technol 50:121–129.
1015
Michud A, Tanttu M, Asaadi S, Ma Y, Netti E, Kaariainen P, Persson A, Berntsson A, Hummel M, Sixta H (2015)
1016
Ioncell-F: ionic liquid-based cellulosic textile fibres as an alternative to viscose and Lyocell. Text Res J 86:543–
1017
552.
1018
Moawia RM, Nasef MM, Mohamed NH, Ripin A (2016) Modification of flax fibres by radiation induced emulsion
1019
graft copolymerization of glycidyl methacrylate. Radiat Phys Chem 122:35–42. doi:
1020
10.1016/j.radphyschem.2016.01.008
1021
Mohanty A.K. Misra M HG (2000) Biofibres, biodegradable polymers and biocomposites: an overview. Macromol
1022
Mater Eng 276/277:1–24.
1023
Mukhopadhyay S, Fangueiro R (2009) Physical modification of natural fibres and thermoplastic films for composites --
1024
A Review. J Thermoplast Compos Mater 22:135–162.
1025
Mwaikambo LY, Ansell MP (2006) Mechanical properties of alkali treated plant fibres and their potential as
1026
Accepted for publication in Industrial Crops and Products on 13 July 2017.
38
reinforcement materials. I. hemp fibres. J Mater Sci 41:2483–2496.
1027
Nabihah S, Jaafar S, Amran UA, Roslan R, Hua CC, Zakaria S (2015) Properties of bio-phenol formaldehyde
1028
composites filled with empty fruit bunch fibre. Int J Chem Mol Nucl Mater Metall Eng 9:171–174.
1029
Naik JB, Mishra S (2006) The compatibilizing effect of maleic anhydride on swelling properties of plant-fibre-
1030
reinforced polystyrene composites. Polym Plast Technol Eng 45:923–927.
1031
Neagu RC (2005) Stiffness Contribution of Various Wood Fibres to Composite Materials. J Compos Mater 40:663–699.
1032
Ni X, Dong A, Fan X, Wang Q, Yu Y, Cavaco-Paulo A (2015) Jute/polypropylene composites: Effect of enzymatic
1033
modification on thermo-mechanical and dynamic mechanical properties. Fibres Polym 16:2276–2283.
1034
Nilsson T, Gustafsson PJ (2007) Influence of dislocations and plasticity on the tensile behaviour of flax and hemp
1035
fibres. Compos Part A Appl Sci Manuf 38:1722–1728.
1036
Nykter M, Kymäläinen H-R, Thomsen AB, Lilholt H, Koponen H, Sjöberg A-M, Thygesen A (2008) Effects of thermal
1037
and enzymatic treatments and harvesting time on the microbial quality and chemical composition of fibre hemp
1038
(Cannabis sativa L.). Biomass and Bioenergy 32:392–399.
1039
Oksman K, Skrifvars M, Selin JF (2003) Natural fibres as reinforcement in polylactic acid (PLA) composites. Compos
1040
Sci Technol 63:1317–1324.
1041
Olaru A, Măluţan T, Ursescu CM, Geba M, Stratulat L (2016) Structural changes in hemp fibres following
1042
temperature , humidity and UV or gamma-ray radiation exposure. Cellul Chem Technol 50:31–39.
1043
Osemeahon SA, Barminas JT, Length F (2007) Study of some physical properties of urea formaldehyde and urea
1044
proparaldehyde copolymer composite for emulsion paint formulation. Int J Phys Sci 2:169–177.
1045
Pakarinen A, Zhang J, Brock T, Maijala P, Viikari L (2012) Enzymatic accessibility of fibre hemp is enhanced by
1046
enzymatic or chemical removal of pectin. Bioresour Technol 107:275–81.
1047
Panaitescu DM, Nicolae CA, Vuluga Z, Vitelaru C, Sanporean CG, Zaharia C, Florea D, Vasilievici G (2016) Influence
1048
of hemp fibres with modified surface on polypropylene composites. J Ind Eng Chem 37:137–146. doi:
1049
10.1016/j.jiec.2016.03.018
1050
Pekala RW, Alviso CT, Kong FM, Hulsey SS (1992) Aerogels derived from multifunctional organic monomers. J Non
1051
Cryst Solids 145:90–98.
1052
Peña MJ, Vergara CE, Carpita NC (2008) The structures and architectures of plant cell walls define dietary fibre
1053
composition and the textures of foods. In: McCleary B V., Prosky L (eds) Advanced Dietary Fibre Technology.
1054
Blackwell Science, pp 42–60
1055
Accepted for publication in Industrial Crops and Products on 13 July 2017.
39
Pepper T (2001) Polyester Resins. ASM Handbook, Compos. 21:90–96.
1056
Perreux D (2012) life prediction of composite materials under complex loading. In: Ifremer-ONR Workshop on the
1057
durability of composites in a matrine environment. pp 75–80
1058
Pickering KL, Beckermann GW, Alam SN, Foreman NJ (2007) Optimising industrial hemp fibre for composites.
1059
Compos Part A Appl Sci Manuf 38:461–468.
1060
Pickering KL, Efendy MGA, Le TM (2015) A review of recent developments in natural fibre composites and their
1061
mechanical performance. Compos Part A Appl Sci Manuf 83:98–112.
1062
Pil L, Bensadoun F, Pariset J, Verpoest I (2016) Why are designers fascinated by flax and hemp fibre composites?
1063
Compos Part A Appl Sci Manuf 83:193–205.
1064
Placet V, Trivaudey F, Cisse O, Gucheret-Retel V, Boubakar ML (2012) Diameter dependence of the apparent tensile
1065
modulus of hemp fibres: A morphological, structural or ultrastructural effect? Compos Part A Appl Sci Manuf
1066
43:275–287.
1067
Pott GT (2002) Reduction of moisture sensitivity in natural fibres. Adv Fibres, Plast Laminates Compos 702:87–98.
1068
Prasad N, Agarwal VK, Sinha S (2016) Banana fibre reinforced low-density polyethylene composites: effect of
1069
chemical treatment and compatibilizer addition. Iran Polym J 25:229–241.
1070
Pritchard G (1998) An introduction to plastics for non-specialists. In: Pritchard G (ed) Reinforced Plastics Durability.
1071
Woodhead Publishing Limited, pp 1–30
1072
Pucciariello R, Villani V, Bonini C, D’Auria M, Vetere T (2004) Physical properties of straw lignin-based polymer
1073
blends. Polymer (Guildf) 45:4159–4169.
1074
Puls J, Janzon R, Saake B (2006) Comparative removal of hemicelluloses from paper pulps using Nitren, cuen, NaOH,
1075
and KOH. Lenzinger Berichte 86:63–70.
1076
Qiu R, Ren X, Fifield LS, Simmons KL, Li K (2011) Hemp-fibre-reinforced unsaturated polyester composites:
1077
optimization of processing and improvement of interfacial adhension. Polym Polym Compos 121:862–868. doi:
1078
10.1002/app
1079
Qiu R, Ren X, Li K (2012) Effect of fibre modification with a novel compatibilizer on the mechanical properties and
1080
water absorption of hemp-fibre-reinforced unsaturated polyester composites. Polym Eng Sci 52:1342–1347. doi:
1081
10.1002/pen
1082
Raghavendra SBS, Vindo B, Sudev LJ (2015) Effect of gamma irradiation on mechanical properties of natural fibres
1083
reinforced hybrid composites. Int J Sci Technol Eng 2:15–23.
1084
Accepted for publication in Industrial Crops and Products on 13 July 2017.
40
Ragoubi M, Bienaimé D, Molina S, George B, Merlin A (2010) Impact of corona treated hemp fibres onto mechanical
1085
properties of polypropylene composites made thereof. Ind Crops Prod 31:344–349.
1086
Ralph J (2010) Hydroxycinnamates in lignification. Phytochem Rev 9:65–83.
1087
Ranalli P, Venturi G (2004) Hemp as a raw material for industrial applications. Euphytica 140:1–6.
1088
Reddy N, Yang Y (2008) Characterizing natural cellulose fibres from velvet leaf (Abutilon theophrasti) stems.
1089
Bioresour Technol 99:2449–2454.
1090
Rosato D V., Rosato MG, Rosato D V. (2000) Concise Encyclopaedia of Plastics. springer
1091
Rouison D, Couturier M, Sain M, MacMillan B, Balcom BJ (2005) Water absorption of hemp fibre/unsaturated
1092
polyester composites. Polym Compos 26:509–525.
1093
Rozman HD, Yeo YS, Tay GS, Abubakar A (2003) The mechanical and physical properties of polyurethane composites
1094
based on rice husk and polyethylene glycol. Polym Test 22:617–623. doi: 10.1016/S0142-9418(02)00165-4
1095
Saint-Michel F, Chazeau L, Cavaillé J-Y, Chabert E (2006) Mechanical properties of high density polyurethane foams:
1096
I. Effect of the density. Compos Sci Technol 66:2700–2708. doi: 10.1016/j.compscitech.2006.03.009
1097
Salentijn EMJ, Zhang Q, Amaducci S, Yang M, Trindade LM (2015) New developments in fibre hemp (Cannabis
1098
sativa L.) breeding. Ind Crops Prod 68:32–41.
1099
Sanadi AR, Prasad S V., Rohatgi PK (1986) Sunhemp fibre-reinforced polyester. J Mater Sci 21:4299–4304.
1100
Sawpan MA, Pickering KL, Fernyhough A (2011a) Effect of various chemical treatments on the fibre structure and
1101
tensile properties of industrial hemp fibres. Compos Part A Appl Sci Manuf 42:888–895.
1102
Sawpan MA, Pickering KL, Fernyhough A (2011b) Effect of fibre treatments on interfacial shear strength of hemp fibre
1103
reinforced polylactide and unsaturated polyester composites. Compos Part A Appl Sci Manuf 42:1189–1196. doi:
1104
10.1016/j.compositesa.2011.05.003
1105
Schutte CL (1994) Environmental durability of glass-fibre composites. Mater Sci Eng R 13:265–324. doi:
1106
10.1016/0927-796X(94)90002-7
1107
Searle TJ, Summerscales J (1999) Review of the durability of marine laminates. In: Pritchard G (ed) Reinforced Plastics
1108
Durability. Woodhead Publishing Limited, pp 219–266
1109
Seki Y, Sarikanat M, Sever K, Erden S, Gulec HA (2010) Effect of the low and radio frequency oxygen plasma
1110
treatment of jute fibre on mechanical properties of jute fibre/polyester composite. Fibres Polym 11:1159–1164.
1111
doi: 10.1007/s12221-010-1159-5
1112
Shah DU, Schubel PJ, Clifford MJ, Licence P (2012) Fatigue characterisation of plant fibre composites for rotor blade
1113
Accepted for publication in Industrial Crops and Products on 13 July 2017.
41
applications. JEC Compos Mag 49:51–54.
1114
Shah DU, Schubel PJ, Clifford MJ, Licence P (2013) Fatigue life evaluation of aligned plant fibre composites through
1115
S-N curves and constant-life diagrams. Compos Sci Technol 74:139–149. doi:
1116
10.1016/j.compscitech.2012.10.015
1117
Shahzad A (2011) Hemp fibre and its composites – a review. J Compos Mater 46:973–986.
1118
Shuda M, Drzal LT, Ray D, Mohanty AK, Mishra M (2008) Natural-fibre composites in the automotive sector. In:
1119
Pickering KL (ed) Properties and Performance of Natural-Fibre Composites. Woodhead Publishing Limited, pp
1120
221–268
1121
Singha AS, Rana RK (2010) Graft Copolymerization of Methyl Methacrylate (MMA) onto Agave americana Fibres and
1122
Evaluation of their Physicochemical Properties. Int J Polym Anal Charact 15:27–42. doi: Pii 918299778
1123
10.1080/10236660903299283
1124
Singha AS, Thakur VK (2008) Effect of fibre loading on urea-formaldehyde matrix based green composites. Iran Polym
1125
J 17:861–873.
1126
Singha AS, Thakur VK (2009) Study of mechanical properties of urea-formaldehyde thermosets reinforced by pine
1127
needle powder. BioResources 4:292–308.
1128
Sreekala MS, Kumaran MG, Joseph S, Jacob M, Thomas S (2000) Oil palm fibre reinforced phenol formaldehyde
1129
composites: influence of fibre surface modifications on the mechanical performance. Appl Compos Mater 7:295–
1130
329.
1131
Staiger M, Tucker N (2008) Natural-fibre composites in structural applications. In: Pickering K (ed) Properties and
1132
Performance of Natural-Fibre Composites. Woodhead Publishing Limited, pp 269–308
1133
Standard A (2014) Standard test method for moisture absorption properties and equilibrium conditioning of polymer
1134
matrix composite materials. ASTM D5229/D5229M-14 1–13. doi: 10.1520/D5229
1135
Stevens MP (1999) Polymer Chemistry: An Introduction, 3rd ed. Oxford University Press, New York
1136
Stuart T, Liu Q, Hughes M, McCall RD, Sharma HSS, Norton A (2006) Structural biocomposites from flax - Part I:
1137
Effect of bio-technical fibre modification on composite properties. Compos Part A Appl Sci Manuf 37:393–404.
1138
Summerscales J (2014) Durability of composites in the marine environment. In: Davies P, Rajapakse YDS (eds)
1139
Durability of Composites in a Marine Environment. Springer US, New York, pp 1–13
1140
Summerscales J, Dissanayake N, Virk A, Hall W (2010a) A review of bast fibres and their composites. Part 2 –
1141
Composites. Compos Part A Appl Sci Manuf 41:1336–1344.
1142
Accepted for publication in Industrial Crops and Products on 13 July 2017.
42
Summerscales J, Dissanayake NPJ, Virk AS, Hall W (2010b) A review of bast fibres and their composites. Part 1 –
1143
Fibres as reinforcements. Compos Part A Appl Sci Manuf 41:1329–1335.
1144
Summerscales J, Grove S (2014) Manufacturing methods for natural fibre composites. In: Hodzic A, Shanks R (eds)
1145
Natural fibre composites: materials, processes and properties. Taylor & Francis, pp 157–186
1146
Summerscales J, Grove S (2013) Manufacturing methods for natural fibre composites. In: Hodzic A, Shanks R (eds)
1147
Natural fibre composites: materials, processess and applications. Woodhead Publishing Limited, pp 176–215
1148
Suwannarangsee S, Arnthong J, Eurwilaichitr L, Champreda V (2014) Production and characterization of multi-
1149
polysaccharide degrading enzymes from Aspergillus aculeatus BCC199 for saccharification of agricultural
1150
residues. J Microbiol Biotechnol 24:1427–1437.
1151
Thomas MR (1983) Isocyanatoethyl methacrylate-a heterofunctional monomer for polyurethane and vinyl polymer
1152
systems.pdf. J Coatings Technol 55:55–61.
1153
Thomassen L V., Larsen DM, Mikkelsen JD, Meyer AS (2011) Definition and characterization of enzymes for maximal
1154
biocatalytic solubilization of prebiotic polysaccharides from potato pulp. Enzyme Microb Technol 49:289–297.
1155
Thomsen AB, Thygesen A, Bohn V, Nielsen KV, Pallesen B, Jørgensen MS (2006) Effects of chemical-physical pre-
1156
treatment processes on hemp fibres for reinforcement of composites and for textiles. Ind Crops Prod 24:113–118.
1157
Threepopnatkul P, Kaerkitcha N, Athipongarporn N (2009) Effect of surface treatment on performance of pineapple
1158
leaf fibre-polycarbonate composites. Compos Part B Eng 40:628–632.
1159
Thuault A, Eve S, Blond D, Bréard J, Gomina M (2013) Effects of the hygrothermal environment on the mechanical
1160
properties of flax fibres. J Compos Mater 48:1699–1707.
1161
Thygesen A (2006) Properties of hemp fibre polymer composites - An optimisation of fibre properties using novel
1162
defibration methods and fibre characterisation. Risø National Laboratory, Roskilde, Denmark
1163
Thygesen A, Daniel G, Lilholt H, Thomsen AB (2006) Hemp fibre microstructure and use of fungal defibration to
1164
obtain fibres for composite materials. J Nat Fibres 2:19–37.
1165
Thygesen A, Liu M, Meyer AS, Daniel G (2013) Hemp fibres: Enzymatic effect of microbial processing on fibre bundle
1166
structure. Risoe Int Symp Mater Sci Proc 34:373–380.
1167
Thygesen A, Madsen B, Bjerre AB, Lilholt H (2011) Cellulosic fibres: Effect of processing on fibre bundle strength. J
1168
Nat Fibres 8:161–175.
1169
Thygesen A, Madsen F, Lilholt H, Felby C, Thomsen A (2002) Changes in chemical composition, degree of
1170
crystallisation and polymerisation of cellulose in hemp fibres caused by pre-treatment. Proc 23th Risø Int Symp
1171
Accepted for publication in Industrial Crops and Products on 13 July 2017.
43
Mater Sci Risø Natl Lab Denmark 315–323.
1172
Thygesen A, Thomsen AB, Daniel G, Lilholt H (2007) Comparison of composites made from fungal defibrated hemp
1173
with composites of traditional hemp yarn. Ind Crops Prod 25:147–159.
1174
Tserki V, Zafeiropoulos NE, Simon F, Panayiotou C (2005) A study of the effect of acetylation and propionylation
1175
surface treatments on natural fibres. Compos Part A Appl Sci Manuf 36:1110–1118. doi:
1176
10.1016/j.compositesa.2005.01.004
1177
Vaca-Garcia C (2008) Biomaterials. In: Clark JH, Deswarte FEI (eds) Introduction to chemicals from biomass. John
1178
Wiley & Sons, Ltd, Padstow, Cornwall, pp 103–141
1179
van der Werf HMG, Mathijssen EWJM, Haverkort a J (1996) The potential of hemp (Cannabis sativa L.) for
1180
sustainable fibre production: A crop physiological appraisal. Ann Appl Biol 129:109–123.
1181
van der Werf HMG, Turunen L (2008) The environmental impacts of the production of hemp and flax textile yarn. Ind
1182
Crops Prod 27:1–10.
1183
van der Werf HMG, van der Veen JEH, Bouma ATM, ten Cate M (1994) Quality of hemp (Cannabis sativa L.) stems
1184
as a raw material for paper. Ind Crops Prod 2:219–227.
1185
van der Werf HMG, van Geel WCA, van Gils LJC, Haverkort AJ (1995) Nitrogen fertilization and row width affect
1186
self-thinning and productivity of fibre hemp (Cannabis sativa L.). Field Crop Res 42:27–37.
1187
van Vuure AW, Baets J, Wouters K, Hendrickx K (2015) Compressive properties of natural fibre composites. Mater
1188
Lett 149:138–140.
1189
Verdu J, Colin X (2012) Humid aging of polymers and organic matrix composites. In: Ifremer-ONR Workshop on the
1190
durability of composites in a matrine environment. pp 27–33
1191
Vignon MR, Dupeyre D, Garcia-Jaldon C (1996) Morphological characterization of steam-exploded hemp fibres and
1192
their utilization in polypropylene-based composites. Bioresour Technol 58:203–215.
1193
Voigt B, Rychwalski RW, McCarthy DMC, Den Adel JC, Marissen R (2003) Carbon fibre reinforced melamine-
1194
formaldehyde. Polym Compos 24:380–390.
1195
Wang H, Xian G, Li H (2015) Grafting of nano-TiO2 onto flax fibres and the enhancement of the mechanical properties
1196
of the flax fibre and flax fibre/epoxy composite. Compos Part A Appl Sci Manuf 76:172–180.
1197
Wang HM, Postle R, Kessler RW, Kessler W (2003) Removing pectin and lignin during chemical processing of hemp
1198
for textile applications. Text Res J 73:664–669. doi: 10.1177/004051750307300802
1199
Yu T, Jiang N, Li Y (2014) Study on short ramie fibre/poly(lactic acid) composites compatibilized by maleic anhydride.
1200
Accepted for publication in Industrial Crops and Products on 13 July 2017.
44
Compos Part A Appl Sci Manuf 64:139–146.
1201
Zelez J (1992) Surface modification of plastic substrates. 1–4.
1202
Zhang J, Zhang H, Zhang J (2014) Evaluation of liquid ammonia treatment on surface characteristics of hemp fibre.
1203
Cellulose 21:569–579.
1204
Zhang LL, Zhu RY, Chen JY, Chen JM, Feng XX (2008) Seawater-retting treatment of hemp and characterization of
1205
bacterial strains involved in the retting process. Process Biochem 43:1195–1201.
1206
Zhang XL, He Y Di (2013) Effects of Laccase on the Properties of Hemp Fabric. Adv Mater Res 690–693:999–1002.
1207
Zhong JB, Lv J, Wei C (2007) Mechanical properties of sisal fibre reinforced ureaformaldehyde resin composites.
1208
Express Polym Lett 1:681–687.
1209
Zini E, Scandola M (2011) Gree composites: an overview. Polym Compos 32:1905–1915. doi: 10.1002/pc
1210
Özmen N (2012) A study of the effect of acetylation on hemp fibres with vinyl acetate. BioResources 7:3800–3809.
1211
1212
1213
Accepted for publication in Industrial Crops and Products on 13 July 2017.
45
Figures
1214
1215
Figure 1. The performance of composites with fibres in a polymer matrix is governed by a range of parameters.
1216
1217
Accepted for publication in Industrial Crops and Products on 13 July 2017.
46
1218
Figure 2. Classification of plant fibres according to their location in the plant.
1219
1220
Accepted for publication in Industrial Crops and Products on 13 July 2017.
47
1221
Figure 3. Schematic diagram of a transverse section of hemp stem showing the organization and morphology of a bast
1222
strip and single fibre (e.g. primary- and secondary fibres) in the bast layer at different levels.
1223
1224
Accepted for publication in Industrial Crops and Products on 13 July 2017.
48
1225
Figure 4. Hemp stem shown at increasing magnification using different transverse sections in SEM. A: Xylem +
1226
cambium + cortex + epidermis; B: Primary and secondary single fibres; C: Major layers in primary single fibre; D: Thin
1227
lamellae within the S2 layer (above) and a model of the microfibril orientation throughout the secondary cell wall
1228
(below) (Thygesen et al. 2006).
1229
1230
Accepted for publication in Industrial Crops and Products on 13 July 2017.
49
1231
Figure 5. SEM micrographs of 0.5% EDTA + 0.2% endopolygalacturonase (EPG) treated fibres (a) and 0.5% EDTA +
1232
0.2% EPG followed by 10% NaOH treated fibres (b) ( 50 µm for images a and b).
1233
1234
Accepted for publication in Industrial Crops and Products on 13 July 2017.
50
1235
Figure 6. Fibre correlated porosity factor versus pectin + hemicellulose content (a) and effective fibre stiffness versus
1236
fibre correlated porosity factor (b) (Liu et al. 2016b).
1237
1238
1239
Accepted for publication in Industrial Crops and Products on 13 July 2017.
51
Tables
1240
Table 1. Desirable properties for fibres to be used as reinforcement agents in unidirectional fiber reinforced composites.
1241
High stiffness and strength
Little scatter in mechanical properties
Long and continuous form of fibres with high specific surface
Low moisture uptake/high moisture resistance
Hydrophobic and smooth fibre surface
Little or no deformation due to changes in temperature or moisture
Other properties e.g. high thermal stability, anti-microbial degradation/microbial decay resistance
1242
1243
Accepted for publication in Industrial Crops and Products on 13 July 2017.
52
Table 2. Advantages and disadvantages of NFCs (Oksman et al. 2003; Faruk et al. 2012; Pickering et al. 2015)
1244
compared to glass fibre reinforced composites.
1245
Advantages
Disadvantages
Production of natural fibres is perceived to have low
environmental impact
Lower durability than synthetic fibre composites,
but durability can be improved with fibre treatments
High specific stiffness and strength
High moisture absorption can result in fibre swelling
Low density
Lower strength and stiffness than synthetic fibres
Lower fibre production costs than synthetic fibres
Considerably higher variability in properties of
fibres
Low hazard manufacturing process
Lower thermal stability
Low emission of toxic fumes when subjected to heat
and during incineration at the end of life
Less stable when subjected to natural
microorganisms because cellulose fibres are
biodegradable
High possibility for modifying fibre surface and
structure
Hydrophilic fibre surface reduces fibre/matrix
interface properties
Composting at end-of-life
Supply chain issues for fibres harvested in the
temperate zone
1246
1247
Accepted for publication in Industrial Crops and Products on 13 July 2017.
53
Table 3 Chemical composition of hemp fibres from different cultivars (hemicellulose refers to pentosans in hemp
1248
fibres)
1249
Cultivar
Cellulose (%)
Hemicellulose (%)
Pectin (%)
Lignin (%)
Reference
USO31
78.4-81.7
5.7-6.4
-
10-13
(Jankauskiene et al. 2015)
Unspecified
58.7
14.2
16.8
6
(Le Troedec et al. 2008)
Fedora 17
65.6-84.9
6.0-8.1
9.4-25
2.7-4.5
(Crônier et al. 2005)
Felina
57.1-61.8
8.3-14.3
2.8-8.6
1.2-7.3
(Thygesen et al. 2002)
Unspecified
76.1-89.2
1.9-12.3
-
2.1-5.7
(Kostic et al. 2008)
Unspecified
82.0-88.9
4.1-8.4
-
2.2-3.8
(Di Candilo et al. 2010)
Unspecified
88.3-91.0
6.5-9.8
-
1.4-2.1
(Madsen et al. 2007b)
Felina 34
64.0-83.0
11.0-15.0
1.0-6.0
1.0-4.0
(Thygesen et al. 2007)
USO
54.4
-
8.7
6.1
(Pakarinen et al. 2012)
Fedora
55
16
8
4
(Bonatti et al. 2004)
Felina
64-76
15
1-6
2-4
(Thygesen et al. 2006)
Unspecified
53.9-57.6
16.3-18.2
6.5-7.1
20-21
(Zhang et al. 2014)
Fibrimon 56
53.2
6.9
-
5.0
(van der Werf et al. 1994)
Fedora 19
58.6
9.3
-
5.0
(van der Werf et al. 1994)
Kompolti
Sárgaszárú
68.2-69.2
6.7-8.5
-
3.5-5.5
(van der Werf et al. 1994)
Kompolti Hybrid
TC
60.2-74.3
7.1-7.9
-
3.3-4.4
(van der Werf et al. 1994)
Unspecified
63.7-78.4
11.8-17.3
1.9-7.3
1.7-5.0
(Thygesen et al. 2011)
1250
1251
Accepted for publication in Industrial Crops and Products on 13 July 2017.
54
Table 4 Mechanical properties of hemp fibre and synthetic fibre.
1252
Fibre type
Density
(g/cm3)
Tensile strength
(MPa)
Stiffness (GP
a)
Failure strain
(%)
Reference
Hemp
1.4-1.5
550-600
25-35
-
(Sawpan et a
l. 2011a)
Hemp
1.5
310-750
30-60
2-4
(Mwaikambo
and Ansell 200
6)
Hemp
368-482
17.6-19.6
2.5-3
(Duval et al.
2011)
Hemp
889 ± 472
35.5 ± 17.3
2.6 ± 2.2
(Marrot et al.
2013)
Hemp
699 ± 450
31.2 ± 19.7
3.3 ± 1.6
(Marrot et al.
2013)
Hemp
489 ± 233
33.8 ± 12.2
2.5 ± 1.3
(Marrot et al.
2013)
Hemp
857 ± 260
58
-
(Pickering et
al. 2007)
Hemp
886
66
-
(Fan et al. 20
11)
Hemp
636 ± 253
27.6 ± 7.5
2.1 ± 0.7
(Placet et al.
2012)
Hemp
1.6
200-1000
-
(Thygesen et
al. 2011)
Glass fibre
2.55
1956
79
-
(Fu et al. 200
0)
Carbon fibre
1.30
3950
238
-
(Fu et al. 200
0)
1253
1254
Accepted for publication in Industrial Crops and Products on 13 July 2017.
55
Table 5 Properties of commonly used thermoplastics in NFCs.
1255
Thermoplastics
Tm*
(°C)
Density**
(g/cm3)
Stiffness
(GPa)
Tensile
strength
(MPa)
Elongation
(%)
References
Example used in
NFCs
Polyester
/
1.2
4
61
2.5
(Brahim
and Cheikh
2007)
(Aziz and Ansell
2004; Seki et al.
2010)
High density
polyethylene (HDPE)
130
0.96
1.0 – 1.1
22 – 31
10 − 1200
(Stevens
1999;
Deanin and
Mead 2007;
Lerche et
al. 2014)
(Pucciariello et
al. 2004; Lu and
Oza 2013)
Low density polyethylene
(LDPE)
108
0.92
0.2−0.3
8.3 – 31
100 – 650
(Stevens
1999;
Deanin and
Mead 2007)
(Pucciariello et
al. 2004; Madsen
et al. 2007a;
Prasad et al.
2016)
Polystyrene (PS)
84 –
106
1.04 –
1.06
2.3 – 3.3
36 – 52
1.2 – 2.5
(Stevens
1999;
Deanin and
Mead 2007)
(Pucciariello et
al. 2004; Naik
and Mishra
2006)
Poly(lactic acid) (PLA)
155
–
171
1.24
3.6
56 − 66
1.2 − 100
(Stevens
1999; Duc
et al. 2014)
(Oksman et al.
2003; Yu et al.
2014)
Poly (vinyl chloride)
(PVC)
199
1.35
2.4 – 4.1
41 – 52
40 – 80
(Stevens
1999;
Rosato et
al. 2000)
(Huang et al.
2012)
Polyamide (Nylon)
260
1.13
/
76 – 83
0.6 − 300
(Stevens
1999;
Rosato et
al. 2000)
(Bourmaud et al.
2016; Kiziltas et
al. 2016)
Polyethyleneterephthalate
(PET)
250
1.34
2.8 – 4.1
48 − 72
50 – 300
(Stevens
1999;
Rosato et
al. 2000)
(Madsen et al.
2007a)
Polypropylene (PP)
168
0.90
1.2 – 1.7
31 – 41
100 − 600
(Stevens
1999;
Rosato et
al. 2000)
(Madsen et al.
2007a)
Polycarbonate (PC)
266
1.20
2.4
66
110
(Stevens
1999;
Rosato et
al. 2000)
(Threepopnatkul
et al. 2009)
Poly (methyl
methacrylate) (PMMA)
95
1.20
2.2 – 3.2
48 − 76
2 − 10
(Stevens
1999;
Rosato et
al. 2000)
(John et al. 2015)
* Tm- melting point of polymers. ** At room temperature.
1256
1257
Accepted for publication in Industrial Crops and Products on 13 July 2017.
56
Table 6 Properties of commonly used thermosets in NFCs.
1258
Thermosets
Curing
temperature
(°C)
Density*
(g/cm3)
Stiffness
(GPa)
Tensile
strength
(MPa)
References
Example used in
NFCs
Epoxy
80 − 150
1.1 – 1.3
2.7 – 4.1
55 − 130
(Crosky et al. 2014; Wang
et al. 2015; Liu et al.
2016c)
(Liu et al.
2016b; Liu et al.
2016c)
Phenol-
formaldehyde
100 − 185
1.6 – 1.2
0.38
10
(Hauptt and Sellers 1994;
Sreekala et al. 2000)
(Sreekala et al.
2000; Nabihah
et al. 2015)
Urea-
formaldehyde
120 − 145
0.9 – 1.6
14
35
(Bliznakov et al. 2000;
Osemeahon et al. 2007;
Zhong et al. 2007; Singha
and Thakur 2008; Singha
and Thakur 2009)
(Zhong et al.
2007; Singha
and Thakur
2008)
Polyester
65 − 80
1.2
0.6 – 3.6
20 − 80
(Sanadi et al. 1986; Pepper
2001; Baley et al. 2006;
Crosky et al. 2014)
(Marais et al.
2005; Dhakal et
al. 2007)
Melamine-
formaldehyde
90 − 120
1.5
9
75
(Pekala et al. 1992; Voigt et
al. 2003)
(Hagstrand and
Oksman 2001)
Polyurethane
80 − 125
1.2
0.43
4.8
(Rozman et al. 2003; Saint-
Michel et al. 2006; Bakare
et al. 2010)
(Hufenbach et
al. 2013)
Furan
40 − 90
1.3
3 − 8.5
30 − 60
(Eisenreich 2008; Džalto et
al. 2014; Domínguez and
Madsen 2015)
(Džalto et al.
2014)
* At room temperature
1259
1260
Accepted for publication in Industrial Crops and Products on 13 July 2017.
57
Table 7 Activities of different enzymes detected in hemp fibres treated by field retting and fungal retting (Liu et al.
1261
2016d).
1262
Enzyme
Activity1 (U/ g dry matter of hemp fibres)
Substrate2
Reference3
Field retting duration
(days)
Fungal retting duration
(days)
7
14
20
7
14
20
Glucanase
0.08
0.10
0.04
0.03
0.07
0.04
10 g/L carboxymethyl
cellulose (CMC)
(Suwannarang
see et al. 2014)
Polygalacturonase
0.04
0.02
0.06
0.60
0.29
0.12
2 g/L polygalacturonic acid
(Thomassen et
al. 2011)
Galactanase
0.11
0.11
0.10
0.70
2.74
0.04
5 g/L potato galactan
(Michalak et
al. 2012)
XG-specific
endoglucanase
0.01
0.08
0.01
0.19
0.32
0.30
10 g/L tamarind xyloglucan
(Benko et al.
2008)
Laccase
0
0
0
0.41
0.48
0.95
0.5 mM ABTS
(Li et al. 2008)
1 The activity of glucanase, polygalacturonase, galactanase and XG-specific endoglucanase: one unit of activity is
1263
defined as the volume of crude enzyme extracts (µL) required to liberate 1 µmole reducing ends (glucose equivalents)
1264
per minute under assay conditions. The activity of laccase: one unit of activity is defined as the volume of crude enzyme
1265
extracts (µL) required to oxidize 1 µmol ABTS per minute under assay conditions.
1266
2ABTS – 2,2-azinobis (3-enthylbenzthiazoline)-6 sulphonate.
1267
3 Reference for methods used for enzyme activity assay in Liu et al. (Liu et al. 2016d).
1268
1269
Accepted for publication in Industrial Crops and Products on 13 July 2017.
58
Table 8 Cell wall components of natural fibres responsible for moisture uptake and biological degradation (adapted
1270
from Mohanty A.K. Misra M 2000).
1271
Component
Moisture uptake
Biological degradation
Hemicellulose
+ + + + + + + +
+ + + + + + + + + + + +
Accessible crystalline cellulose
+ + + + + +
+ + + + + + + + +
Non-crystalline cellulose
+ + +
+ + + + + + + +
Crystalline cellulose
+ +
+ + + + +
Lignin
+
+
* "+" indicates propensity to moisture uptake or to biological degradation.
1272
1273
1274
1275
1276
1277
