![Mechanical properties of natural and synthetic fibre [6-24].](https://www.researchgate.net/profile/Aruan-Mohd-Ghazali/publication/281741104/figure/tbl1/AS:669699083874310@1536680112095/Mechanical-properties-of-natural-and-synthetic-fibre-6-24_Q320.jpg)
Content uploaded by Aruan Efendy Mohd Ghazali
Author content
All content in this area was uploaded by Aruan Efendy Mohd Ghazali on Apr 25, 2016
Content may be subject to copyright.
Review
A review of recent developments in natural fibre composites
and their mechanical performance
K.L. Pickering
a,
⇑
, M.G. Aruan Efendy
a,b
, T.M. Le
a,c
a
School of Engineering, University of Waikato, Hamilton 3216, New Zealand
b
Faculty of Civil Engineering, Universiti Teknologi MARA, Malaysia
c
Department of Textile Technology, Hanoi University of Science and Technology, Hanoi, Viet Nam
article info
Article history:
Available online 9 September 2015
Keyword:
B. Mechanical properties
abstract
Recently, there has been a rapid growth in research and innovation in the natural fibre composite (NFC)
area. Interest is warranted due to the advantages of these materials compared to others, such as synthetic
fibre composites, including low environmental impact and low cost and support their potential across a
wide range of applications. Much effort has gone into increasing their mechanical performance to extend
the capabilities and applications of this group of materials. This review aims to provide an overview of the
factors that affect the mechanical performance of NFCs and details achievements made with them.
Ó2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
Contents
1. Introduction .......................................................................................................... 98
2. Factors affecting mechanical performance of NFCs . . . . ....................................................................... 99
2.1. Fibre selection . . . ................................................................................................ 99
2.2. Matrix selection. . ............................................................................................... 100
2.3. Interface strength ............................................................................................... 101
2.4. Fibre dispersion. . ............................................................................................... 102
2.5. Fibre orientation . ............................................................................................... 103
2.6. Manufacturing . . . ............................................................................................... 103
2.7. Porosity . . . . . . . . ............................................................................................... 104
3. Mechanical properties . . . . . . . . ......................................................................................... 104
4. Hybridisation . . . . . . . . . . . . . . . ......................................................................................... 108
5. Influence of moisture/weathering . . . . . . . . . . . . . . . . . . ...................................................................... 108
6. Applications ......................................................................................................... 109
7. Conclusions. ......................................................................................................... 109
Acknowledgements . . . . . . . . . . ......................................................................................... 109
References . ......................................................................................................... 109
1. Introduction
Interest in NFCs is growing for many reasons including their
potential to replace synthetic fibre reinforced plastics at lower cost
with improved sustainability; their advantages and disadvantages
are summarised in Table 1 [1].
The main factors affecting mechanical performance of NFCs are:
fibre selection – including type, harvest time, extraction
method, aspect ratio, treatment and fibre content,
matrix selection,
interfacial strength,
fibre dispersion,
fibre orientation,
composite manufacturing process and
porosity.
http://dx.doi.org/10.1016/j.compositesa.2015.08.038
1359-835X/Ó2015 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
⇑
Corresponding author. Tel.: +64 78384672.
E-mail address: klp@waikato.ac.nz (K.L. Pickering).
Composites: Part A 83 (2016) 98–112
Contents lists available at ScienceDirect
Composites: Part A
journal homepage: www.elsevier.com/locate/compositesa
Details of the influence of these factors, the mechanical proper-
ties obtained and their applications form the basis for the rest of
this paper.
2. Factors affecting mechanical performance of NFCs
2.1. Fibre selection
Fibre type is commonly categorised based on its origin: plant,
animal or mineral. All plant fibres contain cellulose as their major
structural component, whereas animal fibres mainly consist of pro-
tein. Although mineral-based natural fibres exist within the asbes-
tos group of minerals and were once used extensively in
composites, these are now avoided due to associated health issues
(carcinogenic through inhalation/ingestion) and are banned in
many countries. Generally, much higher strengths and stiffnesses
are obtainable with the higher performance plant fibres than the
readily available animal fibres. An exception to this is silk, which
can have very high strength, but is relatively expensive, has lower
stiffness and is less readily available [6]. This makes plant-based
fibres the most suitable for use in composites with structural
requirements and therefore the focus of this review. Furthermore,
plant fibre can suitably be grown in many countries and can be
harvested after short periods.
Table 2 shows properties of some natural fibres and the main
type of glass fibre (E-glass). It can be seen that flax, hemp and
ramie fibre are amongst the cellulose-based natural fibres having
the highest specific Young’s moduli and tensile strengths although
it should be stated that much variability is seen within the litera-
ture. Commonly, geography relating to fibre availability plays a
major role in fibre selection [7]. The focus, for example in Europe
has been on flax fibre, whereas hemp, jute, ramie, kenaf and sisal
have been of greater interest in Asia. Harakeke fibre, (Phormium
tenax commonly known as New Zealand flax) is also being consid-
ered to be used in structural applications in New Zealand due to its
good mechanical properties and its local availability there.
Generally, higher performance is achieved with varieties having
higher cellulose content and with cellulose microfibrils aligned
more in the fibre direction which tends to occur in bast fibres
(e.g. flax, hemp, kenaf, jute and ramie) that have higher structural
requirements in providing support for the stalk of the plant. The
properties of natural fibres vary considerably depending on chem-
ical composition and structure, which relate to fibre type as well as
growing conditions, harvesting time, extraction method, treatment
and storage procedures. Strength has been seen to reduce by 15%
over 5 days after optimum harvest time [16] and manually
extracted flax fibres have been found to have strength 20% higher
than those extracted mechanically [12]. Strength and stiffness of
natural fibres are generally lower than glass fibre, although stiff-
nesses can be achieved with natural fibres comparable to those
achieved with glass fibre. However, the specific properties com-
pare more favourably; specific Young’s modulus can be higher
for natural fibres and specific tensile strength can compare well
with lower strength E-glass fibres.
When comparing data from different sources, it should be con-
sidered that a number of variables that are not always reported
have an influence on fibre properties. These include testing speed,
gauge length, moisture content and temperature. Generally
strength increases with increasing moisture content and decreases
as temperature increases; the Young’s modulus decreases with
moisture content [25]. Sometimes it is also unclear in the literature
as to whether the tests have been conducted on single fibres (sin-
gle cells) or on fibre bundles (sometimes referred to as technical
fibres). Calculation of properties is generally based on the total
cross-section of a fibre or fibre bundle, however, single fibres have
a central hollow lumen which takes up a significant proportion of
the cross-sectional area. The fraction of cross-sectional area taken
up by the lumen has been found to be, for example, 27.2%, 6.8% and
34.0% for sisal, flax and jute respectively [26] and so it could be
considered that measurements of strength and stiffness obtained
not taking this into account are underestimations to the same
degree. The lumen area fraction for harakeke has been found to
be 41% which based on an apparent strength of 778 MPa obtained
for the fibre neglecting the lumen, suggests a true fibre strength of
1308 MPa [14]. It should also be kept in mind when predicting
composite properties that fibre properties vary with direction
Table 2
Mechanical properties of natural and synthetic fibre [6–24].
Fibre Density
(g/cm
3
)
Length
(mm)
Failure strain
(%)
Tensile strength
(MPa)
Stiffness/Young’s
modulus (GPa)
Specific tensile strength
(MPa/g cm
3
)
Specific Young’s modulus
(GPa/g cm
3
)
Ramie 1.5 900–1200 2.0–3.8 400–938 44–128 270–620 29–85
Flax 1.5 5–900 1.2–3.2 345–1830 27–80 230–1220 18–53
Hemp 1.5 5–55 1.6 550–1110 58–70 370–740 39–47
Jute 1.3–1.5 1.5–120 1.5–1.8 393–800 10–55 300–610 7.1–39
Harakeke 1.3 4–5 4.2–5.8 440–990 14–33 338–761 11–25
Sisal 1.3–1.5 900 2.0–2.5 507–855 9.4–28 362–610 6.7–20
Alfa 1.4 350 1.5–2.4 188–308 18–25 134–220 13–18
Cotton 1.5–1.6 10–60 3.0–10 287–800 5.5–13 190–530 3.7–8.4
Coir 1.2 20–150 15–30 131–220 4–6 110–180 3.3–5
Silk- 1.3 Continuous 15–60 100–1500 5–25 100–1500 4–20
Feather 0.9 10–30 6.9 100–203 3–10 112–226 3.3–11
Wool 1.3 38–152 13.2–35 50–315 2.3–5 38–242 1.8–3.8
E-glass 2.5 Continuous 2.5 2000–3000 70 800–1400 29
Table 1
Advantages and disadvantages of NFCs [2–5].
Advantages Disadvantages
Low density and high specific
strength and stiffness
Lower durability than for
synthetic fibre composites, but
can be improved considerably
with treatment
Fibres are a renewable resource, for
which production requires little
energy, involves CO
2
absorption,
whilst returning oxygen to the
environment
High moisture absorption, which
results in swelling
Fibres can be produced at lower
cost than synthetic fibre
Lower strength, in particular
impact strength compared to
synthetic fibre composites
Low hazard manufacturing
processes
Greater variability of properties
Low emission of toxic fumes when
subjected to heat and during
incineration at end of life
Lower processing temperatures
limiting matrix options
Less abrasive damage to processing
equipment compared with that for
synthetic fibre composites
K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112 99
relative to the fibre axis, although not surprisingly given the exper-
imental challenge, there is only limited information on transverse
data available. The longitudinal Young’s modulus for jute has been
estimated to be seven times that for the transverse Young’s
modulus [27].
As would be expected given that the fibres are normally stron-
ger and stiffer than the matrix, strength and stiffness of the com-
posite are generally seen to increase with increased fibre content.
However, this relies on having reasonable fibre/matrix interfacial
strength, and strength can reduce with strongly hydrophobic
matrices such as polypropylene (PP) with increasing fibre content
unless coupling agents or some other interfacial engineering
method is used; regardless, Young’s modulus still generally
increases with fibre content but more modestly than when the
interface is not optimised [28].
When reasonable interfacial strength is established, composite
strength commonly peaks with fibre contents of 40–55 m% for
injection moulded thermoplastic matrix composites with reduc-
tion at higher contents explained as being due to poor wetting
leading to reduced stress transfer across the fibre–matrix interface
and increasing porosity (see section on porosity below). Stiffness
has been found to increase up to higher fibre contents of around
55–65 m% with similar materials, possibly due to less dependency
on interfacial strength than composite strength [14,28–30]. Fur-
ther insight has been provided by work investigating the influence
of fibre content in terms of weight fraction on porosity and volume
fraction of fibre. This has shown that maximum volume fractions of
fibre occur around fibre contents of 50–60 m% with further addi-
tion of fibre resulting in higher porosity rather than increased fibre
volume fraction, the influence of which has been incorporated into
rule of mixtures models and shown to improve accuracy of predic-
tion for stiffness and strength [27,30,31].
As well as being an issue for short term composite properties,
high fibre volume fractions are also of concern due to the potential
for increased water uptake leading to degradation of longer term
properties. It has been reported that hemp fibre reinforced PP com-
posites with a fibre volume fraction of 0.7 absorbed almost 53 m%
water and had not reached saturation after 19 days, whereas only
7 m% water uptake was observed in composites with a fibre vol-
ume fraction of 0.3 and saturation had been achieved in the same
time period [32].
For composites containing fibres with failure strains lower than
that of the matrix (commonly the case for NFCs), basic composite
theory suggests that there should be a volume fraction of fibre
below which composite strength will be lower than that of the
matrix known as the critical volume fraction (V
crit
). From a fracture
mechanics perspective, below V
crit
, when the fibres fail, the matrix
can cope with load transferred from the failed fibres and the fibres
are acting merely as holes within the matrix. Critical volume frac-
tions of fibre have been found to be 8.1% and 9.3% for jute and flax
respectively in unsaturated polyester (UP), much higher than val-
ues obtained for synthetic fibre composites, although lower than
fibre contents commonly studied in the literature and so this effect
is not observed often [33].
Fibre length, which can be incorporated into the aspect ratio for
a fibre (length/diameter), is an important factor influencing the
mechanical properties of composites. In a short fibre composite,
tensile load is transferred into a fibre from the matrix through
shear at the fibre/matrix interface. At the ends of the fibre, the ten-
sile stress are zero and increase along the fibre length; therefore, a
fibre needs to have a length of greater than a critical length (L
c
)in
order for the fibre to be able to be broken during tensile loading of
a composite [34]. At the critical length, just prior to fracture, the
fibre would theoretically only have been carrying half of the load
compared to that of a continuous fibre at the same composite
strain. Ideally, fibre length would be much greater than the critical
fibre length to allow for efficient reinforcement of a composite
such that the majority of the fibre could be loaded as if it were a
continuous fibre. L
c
can be expressed as follows:
L
c
d¼
r
f
2
s
i
ð1Þ
where dis fibre diameter,
r
f
is tensile strength of fibre and
s
i
is the
interfacial strength.
Not surprisingly, L
c
has been found to vary with fibre, matrix,
fibre treatment and fibre content. L
c
values for hemp/PP compos-
ites with maleated polypropylene (MAPP) coupling agent were
found using composite properties and the Kelly–Tyson model to
be 0.49, 0.67, 0.67 and 0.62 mm for fibre contents of 20, 30, 40,
and 50 m% respectively (
s
i
= 14.5 MPa) [35]. A higher value of
0.83 mm obtained for similar materials was obtained using the
fragmentation test (
s
i
= 15.4 MPa) with alkali treated fibres [29]
with an almost identical value of 0.82 mm (
s
i
= 12 MPa) obtained
for flax with the same matrix based on the Kelly–Tyson model
[36]. Much larger values were obtained from other work using
fragmentation for hemp, flax and cotton with PP and MAPP of
3.2, 3.2 and 5.0 mm (
s
i
= 14.3, 12.0, 0.7 MPa) respectively and
2.3 mm (
s
i
= 22.0 MPa) for sisal [37,38]. Relatively poor bonding
between cotton and PP with MAPP was suggested to be influenced
by lack of lignin which could potentially bond with PP containing
MAPP, which also explains the best interfacial strength being found
for hemp with more accessible lignin at the interface than flax. The
adhesion between polylactic acid (PLA) and the hemp, flax and cot-
ton was found to be insufficient for fragmentation testing analysis
[37].L
c
for flax/thermoset matrix composites has been found to be
generally at the lower end of the range. Values of 0.9, 0.5 and
0.4 mm (and
s
i
= 13, 28 and 33 MPa) for UP, vinyl ester (VE) and
epoxy resins respectively have been obtained using fibre fragmen-
tation supporting their use in NFCs [39]. However, a large L
c
of
3mm (
s
i
= 0.9 MPa) has been observed for jute fibre with UP,
although this was noted as being an upper bound value [40].
Although increasing fibre length generally increases fibre load
bearing efficiency, if fibre length is too long the fibres may get tan-
gled during mixing resulting in poor fibre dispersion which can
reduce the overall reinforcement efficiency [41–43].
2.2. Matrix selection
The matrix is an important part of a fibre-reinforced composite.
It provides a barrier against adverse environments, protects the
surface of the fibres from mechanical abrasion and it transfers load
to fibres. The most common matrices currently used in NFCs are
polymeric as they are light weight and can be processed at low
temperature. Both thermoplastic and thermoset polymers have
been used for matrices with natural fibres [44].
Matrix selection is limited by the temperature at which natural
fibres degrade. Most of the natural fibres used for reinforcement in
natural fibre composite are thermally unstable above 200 °C,
although under some circumstances it is possible for them to be
processed at higher temperature for a short period of time [45].
Due to this limitation, only thermoplastics that soften below this
temperature such as polyethylene (PE), PP, polyolefin, polyvinyl
chloride and polystyrene and thermosets (which can be cured
below this temperature) are useable as a matrix [46]. However, it
should be noted that the thermoplastics named constitute the
most common thermoplastics consumed by the plastics industry
and far outweighs the use of any other thermoplastic matrices gen-
erally used. Indeed, PP and PE are the two most commonly adopted
thermoplastic matrices for NFCs. The main thermosets used are
unsaturated polyester (UP), epoxy resin, phenol formaldehyde
and VE resins. Thermoplastics are capable of being repeatedly soft-
ened by the application of heat and hardened by cooling and have
100 K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112
the potential to be the most easily recycled, which has seen them
most favoured in recent commercial uptake, whereas better reali-
sation of the fibre properties are generally achieved using
thermosets.
Replacement of petroleum-based with bioderived matrices has
been explored. Of these, PLA is the clear front-runner from a
mechanical property perspective, and has been shown to give
higher strength and stiffness with natural fibres than PP [47].
2.3. Interface strength
Interfacial bonding between fibre and matrix plays a vital role
in determining the mechanical properties of composites. Since
stress is transferred between matrix and fibres across the interface,
good interfacial bonding is required to achieve optimum reinforce-
ment, although, it is possible to have an interface that is too strong,
enabling crack propagation which can reduce toughness and
strength. However, for plant based fibre composites there is usu-
ally limited interaction between the hydrophilic fibres and matri-
ces which are commonly hydrophobic leading to poor interfacial
bonding limiting mechanical performance as well as low moisture
resistance affecting long term properties. For bonding to occur,
fibre and matrix must be brought into intimate contact; wettability
can be regarded as an essential precursor to bonding. Insufficient
fibre wetting results in interfacial defects which can act as stress
concentrators [48]. Fibre wettability has been shown to affect the
toughness, tensile and flexural strength of composites [49]. Physi-
cal treatment and chemical treatment can improve the wettability
of the fibre and thus improve the interfacial strength [50–52].
Interfacial bonding can occur by means of mechanisms of
mechanical interlocking, electrostatic bonding, chemical bonding
and inter-diffusion bonding [34]. Mechanical interlocking occurs
to a greater extent when the fibre surface is rough and increases
the interfacial shear strength, but has less influence on the trans-
verse tensile strength. Electrostatic bonding only has significant
influence for metallic interfaces. Chemical bonding occurs when
there are chemical groups on the fibre surface and in the matrix
that can react to form bonds and as a consequence the resulting
interfacial strength depends on the type and density of the bonds.
Chemical bonding can be achieved through the use of a coupling
agent that acts as a bridge between the fibre and matrix. Inter-
diffusion bonding occurs when atoms and molecules of the fibre
and matrix interact at the interface. For polymer interfaces this
can involve polymer chains entanglement and depends on the
length of chains entangled, the degree of entanglement and num-
ber of chains per unit area. It should be noted that it is possible
for multiple types of bonding to occur at the same interface at
the same time [29].
Extensive research has been carried out in order to achieve
improved interfacial bonding in NFCs which can be largely divided
into physical and chemical approaches. Physical approaches
include corona, plasma, ultraviolet (UV), heat treatments electron
radiation and fibre beating. Corona treatment uses plasma gener-
ated by the application of a high voltage to sharp electrode tips
separated by quartz at low temperature and atmospheric pressure
and commonly includes the use of oxygen-containing species [17].
It has been shown to bring about chemical and physical changes of
fibres including increased surface polarity (thought to be due to
increased carboxyl and hydroxyl groups) and increased fibre
roughness but is known to be difficult to apply to three-
dimensional surfaces including fibres [53,54]. Gassan and
Gutowski have used corona plasma and UV to treat jute fibres
which were both found to increase the polarity of fibres but
decrease fibre strength leading to reduced composite strength with
corona treatment, but improvement of up to 30% flexural strength
of epoxy matrix composites with UV treatment [54]. Plasma
treatment is similar to corona treatment but is performed using a
vacuum chamber with gas continuously supplied to maintain the
appropriate pressure and gas composition [17]. Plasma treatment
has been shown to bring about hydrophobicity at fibre surfaces
and increase fibre surface roughness increasing interfacial adhe-
sion [51]. Using plasma treatment, improvement of interlaminar
shear strength and flexural strength in NFCs have been increased
up to 35% and 30% respectively [55]. Heat treatment involves heat-
ing the fibres at temperatures close to those that bring about fibre
degradation and can affect physical, chemical and mechanical
properties of the fibres including water content, chemistry, cellu-
lose crystallinity, degree of polymerisation and strength. Specific
chemical changes include chain scission, free radical production
and formation of carbonyl, carboxyl and peroxide groups [17]. Sim-
ilar to the corona and plasma treatment, the effect of heat treat-
ment relies on time, temperature and composition of the gases
involved during the treatment. Cao et al. obtained improvements
of kenaf fibre tensile strength of over 60% using heat treatment
explained as being due to increased fibre crystallinity [56]. Heat
treatment has also been seen to give good improvement of sisal
fibre strength (37%) [57]. This was associated with removal of aro-
matic impurities as well as increased fibre crystallinity. However,
heat treatment resulted in more modest increases of composite
properties in this work; tensile strength, Young’s modulus, flexural
strength and flexural modulus improvements for composites were
10%, 4%, 27% and 33% respectively. Electron radiation has been seen
to improve interfacial bonding with natural fibres and PP by
between 21% and 53% explained as due to producing free radicals
that encourage crosslinking between the fibre and the matrix
[58]. Fibre beating has brought about a 10% increase in strength
of kraft fibre reinforced PP, which can be explained due to fibre
defibrillation and the associated increased surface area and
mechanical interlocking [59].
Chemical approaches are more represented within the litera-
ture than physical with better improvements obtained to date.
Chemical treatments include alkali, acetyl, silane, benzyl, acryl,
permanganate, peroxide, isocyanate, titanate, zirconate and
acrylonitrile treatments and use of maleated anhydride grafted
coupling agent [47,60]. The most popular are alkali, acetyl, silane
and maleated anhydride grafted coupling agent, but enzyme treat-
ment is becoming increasingly popular with particular benefit
relating to environmental friendliness [47].
Alkali treatment removes fibre constituents including hemi-
cellulose, lignin, pectin, fat and wax which exposes cellulose
and increases surface roughness/area providing for improved
interfacial bonding. Alkali treatment also modifies cellulose
structure; modest treatments have been seen to bring about
increased cellulose crystallinity considered to be due to removal
of materials that could obstruct cellulose crystallinity, whereas
at harsher treatments crystalline cellulose has been converted
to amorphous material [61,62]. Improvement of fibre strength
has also been obtained using alkali treatment [61,63]. Many
studies have reported improvements in interfacial shear strength
(IFSS) and improved tensile strength, Young’s modulus, failure
strain, impact strength, fracture toughness and flexural proper-
ties of composites as well as thermal stability and long term
moisture resistance, the latter of which could be due to the
reduced moisture uptake observed with alkali treated natural
fibres [63–68]. For crystallisable matrices, fibre treatment with
alkali has also been seen to influence the degree of matrix crys-
tallinity, with exposed cellulose acting as a nucleation site for
crystalline polymer [69].
During acetylation of natural fibres, esterification occurs by
reaction of acetyl groups (CH
3
CO–) with hydroxyl groups (–OH)
on the fibres resulting in increased hydrophobicity (see Fig. 1)
[70]. This has been shown to improve interfacial bonding, tensile
K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112 101
and flexural strength and stiffness, as well as dimensional and
thermal stability and resistance to fungal attack in NFCs [71–73].
However, over-treatment has been seen to be deleterious to
mechanical properties, assumed to be due to degradation of cellu-
lose and cracking of fibres known to occur with the catalysts used
in this process [71]. Acetylation treatment has also been found to
reduce impact strength of the composite. Commonly acetylation
is preceded by an alkaline treatment.
Silanes used for treatment of fibres have different functional
groups at either end such that interaction at one end can occur
with hydrophilic groups of the fibre whilst the other end can inter-
act with hydrophobic groups in the matrix to form a bridge
between them. Initially, silane treatment of natural fibres involves
hydrolysis of alkoxy groups on silane with water to form silanol
(Si–OH) groups which can then react with hydroxyl groups on
the fibre surface as shown in Fig. 2 [74,75]; hydrogen or covalent
bonding can occur. The most commonly reported silanes used
are amino, methacryl, glycidoxy and alkyl silanes. Silanes have
been found to increase the hydrophobicity of natural fibres and
strength of NFCs with larger increases occurring when covalent
bonding occurs between silane and the matrix [75,76].
Maleic anhydride (MA) grafted polymers are widely used as
coupling agents to improve composite properties. MA is commonly
grafted to the same polymer as that used as the matrix to ensure
compatibility between the matrix and the coupling agent. MAPP,
produced by grafting MA to PP, is the most commonly seen in
the literature. MAPP can react with the hydroxyl groups on fibre
surfaces leading to covalent or hydrogen bonding Fig. 3. It can be
used as an additive during processing or grafted to the fibre prior
to processing. It has been shown to improve tensile and flexural
strength and stiffness as well as impact strength of PP matrix com-
posites [77]. Of all the methods of improving interfacial bonding,
coupling with MAPP could be regarded as the most successful. It
has been shown to give almost twice the composite strength as
obtained with silane treatment [63]. Further to improved bonding,
improvement of mechanical performance with MAPP has been
explained as being due to its ability to wet fibre and enhance its
dispersion [78]. MA grafted PLA has been shown similarly to
increase properties of PLA matrix NFCs as well as thermal stability
[79–81].
Improvement of composite properties has been seen with the
application of enzyme treatment, with tensile and flexural strength
of abaca/PP composites seen to improve by 45% and 35% respectively
which was considered to be due to removal of fibre components and
increased surface area leading to increased interfacial bonding [82].
Impact properties were also found to increase by 25%.
Further to improvements of instantaneous mechanical perfor-
mance, chemical surface treatment has also been found to improve
longer term mechanical performance of NFCs subjected to wet and
humid conditions [60].
2.4. Fibre dispersion
Fibre dispersion has been identified as a major factor
influencing the properties of short fibre composites and a particu-
lar challenge for NFCs, which commonly have hydrophilic fibres
and hydrophobic matrices [83]. Use of longer fibres can further
increase their tendency to agglomerate. Good fibre dispersion pro-
motes good interfacial bonding, reducing voids by ensuring that
fibres are fully surrounded by the matrix [84]. Dispersion can be
influenced by processing parameters such as temperature and
pressure; additives such as stearic acid have been used in PP and
PE to modify dispersion as well as those used to increase interfacial
bonding such as MAPP which encourage fibre matrix interaction
[83]. Similarly, fibre modification through grafting can also be
employed, but is more expensive [83]. Although the use of more
intensive mixing processes such as using a twin-screw extruder
rather than a single screw extruder leads to better fibre dispersion,
this is generally at the cost of fibre damage and fibre lengths are
found to reduce dramatically during such processing depending
on temperature and screw configuration [61,83].
Fig. 1. Reaction of acetic anhydride with natural fibre.
Fig. 2. Reaction of silane with natural fibre (R
0
representing organic group,...
representing hydrogen bonding).
102 K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112
2.5. Fibre orientation
The best mechanical properties can generally be obtained for
composites when the fibre is aligned parallel to the direction of
the applied load [85–87]. However, it is more difficult to get align-
ment with natural fibres than for continuous synthetic fibres. Some
alignment is achieved during injection moulding, dependent on
matrix viscosity and mould design [88]. However, to get to higher
degrees of fibre alignment, long natural fibre can be carded and
placed manually in sheets prior to matrix impregnation. Alterna-
tively, traditional textile processing of fibres can be employed
including spinning to enable a continuous yarn to be produced.
However, as the name suggests, this does involve a degree of fibre
twist. Aligned fibre yarns can also be produced by wrap spinning, a
method used since the 1970s in the textile industry; here short
fibre can be converted to a continuous form through the use of a
continuous strand wrapped around discontinuous fibre with suffi-
cient frequency to provide the required integrity for subsequent
processing. The continuous strand can be from the same type of
fibre as the short fibre or can be thermoplastic and form the matrix
material during compression moulding (CM). Thermoplastic fibre
to be converted to the matrix can also be used aligned in the yarn
direction to act as support for the natural fibre. Research compar-
ing aligned fibre yarn with conventional twisted yarns for flax
epoxy composites has demonstrated improved tensile and flexural
strength and stiffness with unidirectional yarn [89]. More recently,
continuous fibre tape has been produced through using the fibres
own pectin as an adhesive using a water mist and then drying
whilst stretched [90]. Continuous material can be employed in
processing in much the same way as continuous synthetic fibre
(e.g. filament winding, pultrusion) to give good degrees of fibre
alignment within the composite, although it should not be forgot-
ten that this is not the same as alignment of the cellulose chains as
these tend to have an angle relative to the elemental fibre direction
(microfibrillar angle).
A recently utilised alternative to the textile route with its exten-
sive infrastructure requirements to bring about fibre alignment in
composites is that of dynamic sheet forming (DSF). DSF is a method
used to align fibre traditionally in paper production. Here, short
fibres are suspended in water and sprayed through a nozzle onto
a rotating drum covered with a wire mesh through which the
water can be removed, which brings about alignment in the spray
and rotation direction. This has provided improvements of
mechanical performance compared to other short fibre processing
techniques; recent work at the University of Waikato yet to be
published has obtained strengths of over 100 MPa for short hemp
and harakeke fibre aligned using DSF in PLA and epoxy matrices.
Regarding the degree of influence of orientation on mechanical
performance of NFCs, similar large reductions of strength and
Young’s modulus to those seen with synthetic fibres have been
obtained with increasing fibre orientation angle relative to the test
direction. One study on aligned Alfa fibre reinforced UP showed
strengths compared to those obtained in the fibre direction (0°)
of 69%, 29%, 22% and 12% at angles of 10°,30°,45°and 90°respec-
tively and corresponding Young’s moduli of 93%, 66%, 52% and 41%
of that in the fibre direction [11]. In another study on hemp/PLA
composites, those with fibre aligned at 45°and 90°were found
to have 48% and 30% of the strength and 53% and 42% of the
Young’s moduli of those with 0°fibre [91].
2.6. Manufacturing
The most common methods used for NFCs are extrusion,
injection moulding (IM) and compression moulding. Resin transfer
moulding (RTM) is also used with thermoset matrices and pultru-
sion has been successfully employed for combined flax /PP yarn
composites and thermoset matrix composites [92,93]. Factors
determining properties include temperature, pressure and speed
of processing. It is possible for fibre degradation to occur if the
temperature used is too high, which limits the thermoplastic
matrices used to those with melting points lower than the temper-
ature at which degradation will occur.
In extrusion, thermoplastic, usually in the form of beads or pel-
lets, is softened and mixed with the fibre transported by means of a
single or two rotating screws, compressed and forced out of the
chamber at a steady rate through a die. High screw speed can
result in air entrapment, excessive melt temperatures and fibre
breakage. Low speeds, however, lead to poor mixing and insuffi-
cient wetting of the fibres. This method is used on its own or for
producing pre-cursor for IM. Twin screw systems have been shown
to give better dispersion of fibres and better mechanical perfor-
mance than single screw extruders [94].
IM of composites can be carried out with thermoset or thermo-
plastic matrices, although is much more often used for thermoplas-
tic matrices. Variation of fibre orientation occurs across the mould
section with shear flow along the walls due to friction resulting in
fibre aligned along the mould wall whilst a higher stretch rate at
the centre produces fibre that is more transversely aligned to the
flow direction, a structure referred to as skin core structure [95].
Alignment is more significant with higher fibre contents. Residual
stress in thermoplastic matrix composites due to pressure gradi-
ents, non-uniform temperature profiles, polymer chain alignment
and differences in fibre and matrix thermal expansion coefficients
can reduce composite strength [96]. Due to the viscosity require-
ments, IM of such composites is generally limited to composites
of less than 40 m% fibre content. Fibre attrition in IM as for extru-
sion reduces the length of the fibre during processing.
Compression moulding (CM) is generally used for thermoplastic
matrices with loose chopped fibre or mats of short or long fibre
either randomly oriented or aligned, but can also be used with
thermoset matrices. The fibres are normally stacked alternately
with thermoplastic matrix sheets before pressure and heat are
applied. The viscosity of the matrix during pressing and heating
needs to be carefully controlled, in particular for thick samples to
make sure the matrix is impregnated fully into the space between
fibres. Good quality composites can be produced by controlling vis-
cosity, pressure, holding time, temperature taking account of the
type of fibre and matrix [96]. Film stacking has been recommended
as it limits natural fibre degradation due to involvement of only
one temperature cycle [97]. Temperature still needs to be carefully
controlled as commonly there is little difference in temperature
between that at which a particular matrix can be processed and
that at which fibre degradation will occur. Reduction of fibre
strength has been shown to occur at temperatures as low as
150 °C and at 200 °C, with strength reducing by 10% in ten minutes
[98]. Overall, there is a compromise between obtaining good wet-
ting and avoiding fibre degradation that leads to an optimum tem-
perature for a particular composite material/geometry. This has
been explored with film stacking of flax reinforced poly(ester
amide) composites, with an optimum temperature for composite
tensile properties found to be 150 °C[99]. Flexural properties were
found to be less dependent on temperature below 150 °C, but
reduced significantly at higher temperatures. For composites made
from jute yarn and the bacterial copolyester Biopol
Ò
the optimum
compression temperature for a range of mechanical properties was
found to be approximately 180 °C[1]. The highest strength was
obtained at 200 °C for non-woven mat reinforced PP [100]. Alterna-
tively to film stacking, sheet moulding compounds have been used
in CM [101].
In RTM, liquid thermoset resin is injected into a mould
containing a fibre preform. The main variables with this process
are temperature, injection pressure, resin viscosity, preform
K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112 103
architecture and mould configuration [96]. Advantages compared
with other processes include lower temperature requirements
and avoidance of thermomechanical degradation [102]. Com-
paction in this process is affected by the structure of natural fibres
including the effect of lumen closing and due to lower degrees of
fibre alignment, natural fibre composites are less compactable than
glass fibre composites [102]. Good component strength can be
achieved with this process which is suitable for low production
runs [17].
Processing that is absent from the literature currently, but likely
to be of significance in the future is that of additive fabrication or
3-D printing of NFCs. Early work at the University of Waikato, yet
to be published, shows potential for NFC printing filament to
improve the capabilities of components produced through fused
deposition modelling printers. Here there is benefit of orientation
that occurs when composite material is extruded through a fine
nozzle. It is expected that this will be an area of large growth in
the very near future.
2.7. Porosity
An often overlooked component of NFCs, porosity has long been
known to have shown to have a large influence on mechanical
properties of composites in general and much effort has gone into
reducing it in synthetic fibre composites. It arises due to inclusion
of air during processing, limited wettability of fibres, lumens and
other hollow features within fibres/fibre bundles (which may
become closed during processing at high pressure) and due to
the low ability of fibres to compact [30]. Porosity in NFCs has been
shown to increase with fibre content, more rapidly once the geo-
metrical compaction limit has been exceeded, dependent on fibre
type and orientation of fibre; flax/PP composites were found to
have porosity increasing from 4 to 8 volume% as fibre content
increased from 56 to 72 m% [27]. As mentioned earlier, its inclu-
sion in models has been shown to give improved prediction of
strength and stiffness.
3. Mechanical properties
There is a large amount of literature detailing the mechanical
performance of NFCs. A graphical overview of the range of
strength, stiffnesses, specific stiffnesses and specific strengths
compared with those for glass fibre reinforced plastics produced
recently by Shah [103] is shown in Fig. 4. More details of mechan-
ical properties obtained including tensile and flexural strength and
stiffnesses, as well as impact strength are provided in Table 3,
arranged similarly to Shah’s graph in that more aligned long fibre
composites are given at the top followed by multiaxial, then
short fibre aligned, with random aligned including IM composites
further down, along with regenerated cellulose and glass fibre
reinforced plastics (GFRPs) for comparison at the bottom.
Within each section, thermoset matrix composites are detailed
first. It would appear that similar composite optimisation is
effective for strength and stiffness; Shah found strength as well
as specific strength to be largely proportional to stiffness and
specific stiffness respectively for NFCs as well as the trends for rel-
ative strength and stiffness maintained for specific strength and
stiffness [103].
It can be seen that alignment of fibre is a major factor influenc-
ing composite properties, with the best tensile, flexural and impact
properties achieved for aligned NFCs. The highest strength is
achieved at approximately 73 m% fibre for aligned fibre compos-
ites, a higher fibre content than that from which the best strengths
can be achieved with more randomly aligned/shorter fibre
composites, presumably due to the higher compaction limit with
more aligned fibre. Interestingly, the fibre with which the highest
tensile strength was achieved is sisal, which from Table 2, would
not be expected to bring about such high composite strengths as
flax. This suggests that either the properties in Table 2 do not fully
reflect the spectrum achievable with natural fibres, which would
not be surprising given the limited data available and/or that the
strength of the fibre is of less influence than other factors such as
its aspect ratio, extraction method, treatment or ability to be
aligned within the composite and manufacturing method. Compar-
ison of composites reinforced using wrap spun yarn with conven-
tionally twisted yarn have shown better flexural properties using
discontinuous hemp with PP fibre as a carrier fibre, wrap spun
by PP as well as flax sliver with PP fibre also wrap spun using PP
for unidirectional composites [104]. Recent work using wrap spun
hemp/PLA yarn from bleached hemp yarn and continuous PLA
strand to produce oriented prepreg (pre-impregnated fibre) has
given a reasonable combination of mechanical properties for
biaxial laminates including relatively high Charpy impact energy
(25 kJ/m
2
)[105]. The same group extended wrap spinning of PLA
to produce yarn from short hemp fibre with promising results for
this technique [91]. Although it is early days in the development
of such material for composites, it would be expected that this
approach will produce high mechanical performance.
It should be realised that NFC composites with aligned fibres
either are produced by hand if the fibre is in a reasonably raw form
or require textile infrastructure or alternative processing to
produce a continuous fibre form. Although yarn and sliver give
good performance, processing is particularly intensive with many
stages including skutching, carding or hackling and spinning for
which each stage requiring specialist equipment [106,107].
Relatively high values (136 MPa with epoxy and 101 MPa with
PLA) have been obtained at Waikato University with short fibres
aligned using dynamic sheet forming; these overlap at the lower
end of the strength range for those obtained using long aligned
fibre composites, but not achieving the higher strengths obtained
with longer fibre. Fibre length is shown to be important for more
randomly arranged fibre composites with those having long fibre
performing better than those with shorter fibre. Generally the
highest tensile and flexural properties for NFCs are achieved with
Fig. 3. Interaction of maleated polypropylene with natural fibre (...representing hydrogen bonding).
104 K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112
thermoset matrices for which the highest values in descending
order are for epoxy, vinyl ester then unsaturated polyester, relating
to the order of expected degree of adhesion with the fibre and
hence stress transfer capability. However, it should not be over-
looked that very high strengths are achieved with PP (321 MPa)
and PLA (223 MPa) with flax and kenaf respectively, which appears
to be due to good quality fibre and composite processing as well as
a high fibre content. The best tensile strength for random/IM NFCs
have a PLA matrix, however, it should be noted that thermosets are
less commonly used for these composites due to their higher cost
and the lower performance expected. PLA quite consistently out-
performs PP as a matrix material, not surprisingly given its higher
properties, supporting this bioderived material for application.
Another bioderived matrix, shellac, can also be seen to compare
favourably [98]. However, flexural strength as high as 188 MPa
with randomly aligned alkali treated fibre can be seen in Table 3
for harakeke fibre with epoxy resin, supporting the use of vacuum
bagging followed by CM for quality manufacture. RTM is also
shown to be able to be used to produce high performing NFCs;
the highest stiffness in Table 3 (39 GPa) is produced this way, how-
ever this work also supports the importance of fibre quality with
biotechnical retting employed with fibre produced in a northern
climate with long daylight growing hours noted for their contribu-
tion to performance [108]. It has been raised that the highest fibre
and lowest void contents occur with pre-pregging of fibre followed
by compression moulding, then RTM and hand-lay-up which
matches a trend in reducing tensile properties [103]. Important
processing details raised by Phillips et al. are the need for evacua-
tion of moisture during cure of thermoset matrix composites and
reduction of crimp [109]. High strength and stiffness for random
film stacked flax fibre has been explained by high fibre strength
(1339 MPa) along with processing facilitating good interfacial
bonding whilst avoiding fibre degradation [97].
MAPP has been used as a coupling agent much more commonly
with short rather than long fibre PP matrix composites to bring
about improved mechanical performance, presumably due to the
increased requirement for interfacial stress transfer with short
fibre composites. The greatest impact resistant composites are
those with PP matrices for which, as for other properties, long fibre
NFCs have the highest values. Impact toughness has been found to
reduce with addition of natural fibre to tough thermoplastic matri-
ces such as PP, although less so with addition of MAPP, but has
been seen to improve with addition of natural fibre to PLA and
PHB [110,111]. Charpy impact has been seen to increase for UP
almost linearly with increased fibre content [112]. Untreated sisal
fibre has been shown to give a ten-fold increase in Charpy impact
strength compared to UP alone with acrylamide and permanganate
fibre treatments improving Charpy impact strength, but acetyla-
tion, alkali and silane treatments having a negative influence
[113]. The impact strength of epoxy resin has been seen to reduce
with addition of fibre up to 25 m% fibre, but then increase to give
an overall improvement in impact strength of 40% at a fibre
content of 40 m% [114].
Although impact toughness gives an indication of the tendency
for a material to behave in a brittle manner, plane strain fracture
toughness (k
ic
) is required to prevent brittle fracture. Nonetheless,
data is quite scarce for k
ic
of NFCs. k
ic
of injection moulded short
hemp fibre composites was found to decrease with fibre content
from 2 MPa m
1/2
for just PLA to 1.6 MPa m
1/2
with 30 m% fibre
[69]. Alkali fibre treatment brought about further reduction of k
ic
to 1 MPa m
1/2
. Reduction of k
ic
also occurred for long hemp fibre/
PLA composites in the same work, but to less of a degree than with
shorter fibre. However in contrast, k
ic
values increased for PLA with
addition of long hemp fibre from 2 MPa m
1/2
to 3.3 MPa m
1/2
for
untreated fibre and 3.9 MPa m
1/2
for alkali treated long fibre [67].
From these contradictory results, it seems possible that the
Fig. 4. Comparison of stiffness (tensile modulus), strength and specific stiffness (tensile modulus/density) and specific strength (strength/density) of NFCs (upper of paired
bars) with glass fibre reinforced plastics (lower of paired bars) reprinted from Materials and Design, 62, Darshil U. Shah, Natural fibre composites: Comprehensive Ashby-type
materials selection charts, 21–31, 2014, with permission from Elsevier. (For interpretation of the references to colour in this figure legend, the reader is referred to the web
version of this article.)
K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112 105
Table 3
Mechanical properties of natural fibre composites compared with regenerated cellulose composites and GFRPs.
Fibre Matrix Fibre content
(m%)
Tensile
strength (MPa)
Stiffness/Young’s
modulus (GPa)
Flexural
strength (MPa)
Flexural
modulus (GPa)
Charpy (c) or Izod (i)
impact strength (kJ/m
2
or J/m)
Notes: Processing/length/
treatment
Reference
Sisal (aligned) Epoxy 73 410 6 320 27 Alkali treated bundles CM/leaky
mould
[57]
Sisal (aligned) Epoxy 77 330 10 290 22 Untreated bundles CM/leaky mould [57]
Flax (aligned) Epoxy 46/54 280/279 35/39 Enzyme extracted RTM [108]
Harakeke
(aligned)
Epoxy 50/55 223 17 223 14 CM [14]
Harakeke
(aligned)
Epoxy 52 211 15 CM [115]
Sisal (aligned) Epoxy 48 211 20 RTM [116]
Sisal (aligned) Epoxy 37 183 15 RTM [116]
Flax (yarn)
(aligned)
Epoxy 45 311 25 Not stated [106]
Hemp (aligned) Epoxy 65 165 17 180 9 15 (c) CM [117]
Flax yarn (aligned) Epoxy 31 160 15 190 15 Hand lay–up (knitted yarn) [118]
Flax yarn (aligned) Epoxy 45 133 28 218 18 Autoclave [106]
Flax (aligned) Epoxy 37 132 15 RTM [116]
Flax hackled
(aligned)
Epoxy 28 182 20 Pultruded [118]
Flax yarn (aligned) VE 24 248 24 RTM [118]
Flax (sliver)
(aligned)
UP 58 304 30 Soxhlet extracted Vacuum
impregnated/CM
[107]
Flax yarn (aligned) UP 34 143 14 198 17 RTM (knitted yarn) [118]
Alfa (aligned) UP 48 149 12 Alkali treated then bleached [11]
Flax yarn (aligned) PP 72 321 29 Filament wound [27]
Flax yarn (aligned) PP 30 89/70 7/6 88/115 (c) Pultruded flax/PP yarn [92]
Flax (aligned) PP 50 40 7 751 (i) Needle punched flax/PP mats CM [119]
Flax (aligned) PP 39 212 23 Dew retted, boiled, MAA-PP coupled [100]
Flax sliver
(aligned)
PP 44 146 15 Wrap spun flax sliver/PP hybrid yarn,
CM
[104]
Hemp (aligned) PP 46 127 11 Wrap spun, short hemp/PP hybrid
yarn, CM
[104]
Kenaf selected
(aligned)
PLA 80 223 23 254 22 Emulsion PLA Prepreg CM [120]
Hemp (carded) PLA 30 83 11 143 7 9 (c) Alkali treated CM [67]
Kenaf (aligned) PLA 40 82 8 126 7 14 (c) CM [111]
Hemp (aligned) PLA 30 77 10 101 7 19 (c) Wrap spun alkali treated short hemp
hybrid yarn, CM
[91]
Kenaf (aligned) PHB 40 70 6 101 7 10 (c) CM [111]
Flax sliver
biaxial/major
axis
Epoxy 46 200 17 194 13 Wrap spun sliver, woven, weft:warp
strength 10:1
[89]
Flax (woven) Epoxy 50 104 10 Sized and dried prior to pre-preg [109]
Flax yarn (woven) VE 35 111 10 128 10 RTM [118]
Jute (woven) UP 35 50 8 103 7 11 (c) RTM [93]
Hemp (biaxial) PLA 45 62 7 124 9 25 (c) Wrap spun bleached hemp hybrid
yarn, CM
[105]
Harakeke (DSF) Epoxy 45 136 11 155 10 10 (c) Alkali treated CM Waikato
Hemp (DSF) Epoxy 50 105 9 126 8 Alkali treated CM Waikato
Hemp (DSF) Epoxy 65 113 18 145 10 11 (c) CM [117]
Harakeke (DSF) PLA 30 102 8 Alkali treated CM Waikato
Hemp (DSF) PLA 25 87 9 Alkali treated CM Waikato
106 K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112
Flax (short-non
woven)
Shellac 49 109 10 [98]
Harakeke
(random)
Epoxy 45 188 9 Alkali treated Vacuum bagged CM [121]
Flax (random) UP 39 61 6 91 5 13 (c) RTM [93]
PALF (random) UP 30 53 2 80 3 24 (c) CM [43]
Wood BKP PP 40 50 3 78 3 40 (i) MAPP coupled IM [122]
Flax PP 30 74
b
22
c
(c) MAPP coupled IM [123]
Jute PP 60 74 11 112 12 195 (i) MAPP coupled IM [124]
Newsprint PP 40 53 3 94 4 200 (i) MAPP coupled IM [125]
Kraft PP 40 52 3 90 4 235 (i) MAPP coupled IM [125]
Hemp PP 40 52 4 86 4 210 (i) MAPP coupled IM [125]
Kenaf (random) PP 30 46 5 58 4 39 (i) CM [126]
Flax PP 30 52 5 60 5 18 (c) IM [71]
Flax (random) PLA 30 100 8 Dew retted Stripped/combed
(strength 1339 MPa) Film stacking
[97]
Flax (random) PLLA 30 99 9 Dew retted, stripped, combed
(strength 1339 MPa) Film stacking
[97]
Hemp (random) PLA 47 55 9 113 [127]
Cellulose
(continuous)
Bio-
Epoxy
92 9 727 27 26 (c) RTM [128]
Cordenka
a
PA 30 120 6 IM [129]
Cordenka
a
PP42904––87(c) MAPP coupled IM [130]
Lyocell
a
(carded) PLA 30 89 9 148 6 52 (c) CM [111]
Cordenka
a
PLA251084––70(c) IM [130]
Lyocell
a
(carded) PHB 30 66 5 105 5 70 (c) CM [111]
E-glass
(unidirectional)
VE 60 905 39 [118]
E-glass (aligned) UP 60 695 31 CM [107]
E-glass (aligned) Epoxy 41 450 21 Pultruded [118]
E-glass (woven) VE 59 483 33 [118]
E-glass (CSM) Epoxy 38–40 250 9 [121]
E-glass (CSM) UP 47 201 13 278 11 107 (c) RTM [93]
BKP = bleached kraft pulp, PALF = pineapple leaf fibre.
PHB = poly(3-hydroxybutyrate), PLLA =
L
-polylactide acid, PA = Polyamide.
MAA-PP = maleic acid anhydride modified PP.
CSM = chopped strand mat.
a
Lyocell/Cordenka = regenerated cellulose fibre.
b
High molecular weight MAPP
c
Rubbery MAPP.
K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112 107
influence of treatment and fibre length are large influences on the
fracture mechanics and plane strain fracture toughness.
Regenerated cellulose fibre composites (cellulose, Lyocell and
Cordenka) can be seen in Table 3 to have useful properties. Indeed,
properties for IM material generally surpass values seen with IM
NFCs. Impact strength of 72 kJ/m
2
obtained for 30 m% cordenka
fibre reinforced PLA is reported as the highest value seen in the lit-
erature for any bio-degradable composite [131]. Improvements of
impact strength and tensile strength were also obtained by Bledzki
et al. [132] for composites made from PLA with abaca and synthetic
cellulose using IM; notched Charpy impact strength of composites
with 30 m% abaca and cellulose fibres were improved by 120%
(5.7 kJ/m
2
) and 360% (7.9 kJ/m
2
) compared to PLA alone. Compar-
ison of hybrid hemp/Lyocell fibre composites with hemp or Lyocell
only fibre composites showed improvement of tensile and impact
strengths but reduction of stiffness [133]. The flexural strength of
727 MPa obtained with RTM cellulose bio-epoxy is outstanding,
however, the tensile properties are more modest. Use of regener-
ated fibres has the benefit of consistent quality.
Due to the advantages of NFC given in Table 1, there is a strong
driving force to use them to replace GFRPs. From Fig. 4, it can be
seen that stiffness (tensile modulus) values for NFCs approach
the upper values obtained for GFRPs, although strength falls well
short. Taking account of density provides a more favourable com-
parison; higher specific stiffnesses can be achieved for NFCs than
GFRPs, but the highest specific strengths are obtained with GRRPs.
There is though good overlap of specific strength for NFCs and
GFRPs with thermoplastic matrices. A number of studies have
involved comparing GFRPs with NFCs made using the same proce-
dures. Specific stiffness of flax fibre reinforced PLLA composites
were found to be similar to those for glass fibre reinforced UP for
random aligned film stacked composites, but specific strength
was 20% higher for GFRPs [97]. Unidirectional (yarn or sliver) hemp
fibre reinforced SCONACELL A (starch based matrix) laminates
have been found to have 143% of the specific stiffness of glass fibre
reinforced epoxy composites at similar fibre volume fraction (0.5)
although only 59% of their specific strength [98]. Flexural modulus
was found to be almost similar for flax/epoxy composites to epoxy
reinforced with glass CSM at the same weight fraction but the flex-
ural strength was only about two-thirds that for the GFRPs [121].
Impact strength is considered to be one of the weaknesses of
NFCs [47]. The highest Charpy impact energy found in the
literature is for unidirectional pultruded flax/PP composites of
115 kJ/m
2
[92]. However, based on relatively low flexural proper-
ties (strength/modulus 70 MPa/6 GPa) for the same composites, it
was assumed that this was due to poor impregnation of fibre. Pro-
cessing aimed at improving impregnation improved the flexural
properties by around 15%, but reduced the Charpy impact energy
to around 35 kJ/m
2
. Higher die temperature combined with faster
pulling speed lead to a composite more rounded in terms of
mechanical properties with flexural strength, flexural modulus
and Charpy impact strength of 89 MPa, 7 GPa and 88 kJ/m
2
respec-
tively. The highest Izod impact strength (751 J/m) has been
achieved with needle punched flax/PP compression moulded mat
in the 0°direction [119].
Overall, it has been stated that composites reinforced with
natural fibres compare well with GFRPs with respect to stiffness
and cost, but not for impact and tensile strength or water
absorption [93].
Development of nanocellulose fibre reinforced composites has
been inspired by the high predicted mechanical properties of
nanocellulose. Although strength and stiffness of nanocellulose
have not been directly measured, predictions for strength range
from 300 MPa to 23 GPa (fragmentation experiments support
strengths between 1.3 and 1.6 GPa) and stiffness from 58 to
300 GPa [134]. Sources of nanocellulose are plentiful including
the same sources from which macroscopic cellulose based fibres
described already are obtained, as well as animals, algae and bac-
teria. However, processing is a challenge for these fibres with
extraction generally involving multiple stages resulting in fibre
suspensions, the use of which needs to avoid fibre aggregation
and generally limits the composite matrices used [17]. Nanocellu-
lose composite strengths and stiffnesses of up to 420 MPa and
21 GPa respectively have been reported based on use of bacterial
nanocellulose, which puts them at the upper end of what can be
achieved with NFCs [134].
4. Hybridisation
Recent studies have yielded promising results with hybridiza-
tion of natural fibres for reinforcement. The variation in mechani-
cal properties such as tensile, flexural and impact strength of
hybridized kenaf and pineapple leaf fibre reinforced high-density
polyethylene (HDPE) composites has been studied [135]; it was
found that pineapple leaf fibre increased the tensile and flexural
strength whilst kenaf improved impact strength and reduced
water absorption. Evaluation of the effect of hybridization on
mechanical performance of short banana/sisal hybrid fibre rein-
forced polyester composites found that tensile properties of NFCs
were improved with addition of banana fibres [136]. The maxi-
mum tensile strength (58 MPa) was obtained for composites hav-
ing a ratio of banana:sisal of 3:1 at 67 vol% fibre content. The
results were explained as being due to the smaller diameters of
banana fibres compared to sisal fibres and better stress transfer
in unit area of banana/polyester composite.
5. Influence of moisture/weathering
As mentioned previously, moisture absorption is one of the
main disadvantages experienced with NFCs. It has been shown to
increase with increased fibre content and temperature as well as
being influenced by fibre treatment/coupling agent and fibre
arrangement. It is commonly associated with swelling of NFCs
and reduced mechanical performance with the exception of impact
energy which is commonly seen to increase.
Even with hydrophobic matrices such as PP, tensile and flexural
properties of NFCs have been found to reduce considerably over a
few weeks submersion in water with degradation rate increasing
at higher temperatures [32,137]. MAPP used as a coupling agent
has been seen to slow moisture uptake, reduce saturation moisture
content and provide better properties after exposure to moisture
for reinforced PP matrix composites [59,137]; however, large
reductions in properties even with MAPP are found to occur, with
for example reductions of strength for wood/PP composites after
238 days, during which time saturation moisture content occurred,
of 32%, 45% and 59% for 30, 50 and 70 °C respectively.
In work on hemp fibre reinforced PLA, non-woven fibre com-
posites have been found to be more absorbent than aligned fibre
composites; this was considered to be due to increased porosity
due to the increased complexity of the matrix flow path during
processing [91]. Alkali treatment was found to increase resistance
to moisture in the same work. Saturation of moisture has been
found to occur after about two months for aligned hemp reinforced
PLA with reductions of strength (13%), stiffness (22%) and fracture
toughness (25%) but a large increase of impact energy of 530% at
25 °C with larger reductions of strength (30%), stiffness (30%) and
fracture toughness (68%) and increase of impact energy (550%) at
50 °C all for alkali treated fibre [138].
Moisture absorption has been found to be different depending
on its salt content. Absorption rates for composites containing cur-
aua and E-glass fibres with an UP matrix for over 330 h of exposure
108 K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112
to sea water have been seen to be slower than for exposure to dis-
tilled water, suggested to be due to NaCl ions migrating to fibre
surfaces [139]. Reduced diffusion of sea water relative to distilled
water in composites is further supported by work with jute and
glass UP composites [140].
Although the presence of natural fibres is generally considered
to increase moisture absorption of polymers, benefit of inclusion of
natural fibres in glass fibre reinforced thermoset composites has
been found for prolonged moisture exposure; although degrada-
tion rate with jute fibres was found to be faster for times of less
than 70 h, above this time, jute was found to reduce degradation
rate. This has been explained as the swollen jute fibre layers being
able to accommodate the resin swelling strain or just acting as pro-
tection to a central glass fibre layer [141]. In this same research,
although silane coupling was found to improve flexural strength,
the benefit was lost after nearly 12 days in boiling water.
Weathering has, however, highlighted the benefit of natural
fibres. Assessment of mechanical and thermal stability of date
palm leaf fibres in PP composites containing UV stabiliser exposed
to natural weathering conditions in Saudi Arabia and accelerated
weathering conditions utilising UV wavelengths between 300
and 400 nm at 50% humidity, has shown the composites to be more
stable than PP without natural fibre for natural and accelerated
weathering conditions, although MAPP was found to reduce stabil-
ity [142]. It was suggested that lignin acting as a natural antioxi-
dant and it darkening on exposure, increases protection from UV,
as well as enhanced adhesion due to oxidation of the matrix pro-
vided stability in the composites.
6. Applications
Over the last couple of decades, increasing numbers of car mod-
els, first in Europe encouraged by government legislation and then
in North America, have featured natural fibre-reinforced polymers
in door panels, package trays, hat racks, instruments panels, inter-
nal engine covers, sun visors, boot liners, oil/air filters progressing
to more structurally demanding components such as seat backs
and exterior underfloor panelling. Now, all of the main interna-
tional automotive manufactures use these materials and their
use is expected to increase in this area [47]. In India composite
board has been developed as an alternative to medium density
fibreboard which has been assessed for use in railcars [17]. The air-
craft industry has also been adopting NFCs for interior panelling
[96]. They have been used in applications as diverse as toys, funeral
articles, packaging, marine railings and cases for electronic devices
such as laptops and mobile phones as a replacement for synthetic
fibre [96]. In sports, surfing appears to be of particular note with
respect to embracing environmentally friendly materials. A num-
ber of companies are now advertising surfboards incorporating
NFCs. One of the earliest was the ‘‘Ecoboard” produced by Lamina-
tions Ltd, using bio-based resin and hemp fibre [17]. A recent study
supported the production of natural fibre surfboard fins through
RTM to be possible regarding provision of mechanical performance
and economic viability [143]. Fishing rods are also being produced
using material developed by CelluComp Ltd who extract nanocellu-
lose from root vegetables [17,144]. Recent research showing RTM
flax reinforced polyester turbine blades to be a potential replace-
ment for those reinforced with glass fibre obtained recognition
by way of the Asia 2013 Innovation Award from the JEC composites
group for the world’s first functional flax composite wind turbine
blade [145]. Research has also shown the potential of NFCs for
top-plates of string musical instruments [146].
In the construction industry, wood fibre/PP or fibre/PE has
been used extensively in decking, particularly in the USA. Natural
fibre reinforced composites have also been gaining popularity in
non-structural construction applications and used for door and
window frames, 21 wall insulation and floor lamination [147–
150]. Assessment for replacement of wooden laminates in insulat-
ing structural panels have found NFCs to have better mechanical
properties [151]. The possibility of using natural fibre composite
sheet piles by evaluating the flexural behaviour of extruded hollow
cross-section wood-plastic composites (WPCs) with 50 m% wood
flour has been investigated [152]. Results highlighted significant
promise for natural fibre light duty sheet piling structures to
replace conventional sheet piles made from concrete and steel.
There has also been suggestion of similar materials to be used for
beams and slabs [153]. Reinforcement of cement by natural fibres
for building materials is also being assessed [47].
Overall, the global NFCs market was estimated at US$2.1 billion
in 2010 and projected to rise 10% annually until 2016 [96] reflect-
ing further potential seen across a range of industries including
automotive, aerospace, construction, civil and the sports and lei-
sure industries.
7. Conclusions
Much research and progress has occurred in recent decades in
the mechanical performance of NFCs. Improvement has occurred
due to improved fibre selection, extraction, treatment and interfa-
cial engineering as well as composite processing. This paper has
reviewed the research that has focussed on improving strength,
stiffness and impact strength including the effect of moisture and
weathering on these properties; long and short term performance
was addressed. NFCs now compare favourably with GFRPs in terms
of stiffness and cost; values of tensile and impact strength are
approaching those for GFRFs. The lower densities for NFCs lead
to better comparison for specific properties. Applications of NFCs
have extended dramatically including load bearing and outdoor
applications such as automotive exterior underfloor panelling,
sports equipment and marine structures. Further research is still
required to extend their application range including improvement
of moisture resistance and fire retardance. Overall, growth of NFC
uptake continues at a rapid rate and there would appear to be a
very positive future ahead for their application.
Acknowledgements
The authors would like to thank Janine Williams and Cheryl
Ward for their support in producing this review.
References
[1] Mohanty AK, Khan MA, Sahoo S, Hinrichsen G. Effect of chemical modification
on the performance of biodegradable jute yarn-Biopol (R) composites. J Mater
Sci 2000;35(10):2589–95.
[2] Cao Y, Wu Y. Evaluation of statistical strength of bamboo fiber and
mechanical properties of fiber reinforced green composites. J Cent South
Univ Technol 2008;15:564–7.
[3] Lee BH, Kim HJ, Yu WR. Fabrication of long and discontinuous natural fiber
reinforced polypropylene biocomposites and their mechanical properties.
Fibers Polym 2009;10(1):83–90.
[4] Li X, Tabil LG, Panigrahi S. Chemical treatments of natural fiber for use in
natural fiber-reinforced composites: a review. J Polym Environ 2007;15
(1):25–33.
[5] Mehta G, Mohanty AK, Thayer K, Misra M, Drzal LT. Novel biocomposites
sheet molding compounds for low cost housing panel applications. J Polym
Environ 2005;13(2):169–75.
[6] Shah DU, Porter D, Vollrath F. Can silk become an effective reinforcing fibre? A
property comparison with flax and glass reinforced composites. Compos Sci
Technol 2014;101:173–83.
[7] Mustafa A, Bin Abdollah MF, Shuhimi FF, Ismail N, Amiruddin H, Umehara N.
Selection and verification of kenaf fibres as an alternative friction material
using Weighted Decision Matrix method. Mater Des 2015;67:577–82.
[8] De Rosa IM, Kenny JM, Puglia D, Santulli C, Sarasini F. Tensile behavior of New
Zealand flax (Phormium tenax) fibers. J Reinf Plast Compos 2010;29(23):
3450–4.
K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112 109
[9] Dittenber DB, GangaRao HVS. Critical review of recent publications on use of
natural composites in infrastructure. Composites Part A 2011;43(8):1419–29.
[10] Zini E, Scandola M. Green composites: an overview. Polym Compos 2011;32
(12):1905–15.
[11] Brahim SB, Cheikh RB. Influence of fibre orientation and volume fraction on
the tensile properties of unidirectional Alfa-polyester composite. Compos Sci
Technol 2007;67(1):140–7.
[12] Bos HL, Van den Oever MJA, Peters O. Tensile and compressive properties of
flax fibres for natural fibre reinforced composites. J Mater Sci 2002;37
(8):1683–92.
[13] Reddy N, Jiang QR, Yang YQ. Biocompatible natural silk fibers from Argema
mittrei. J Biobased Mater Bioenergy 2012;6(5):558–63.
[14] Le MT, Pickering KL. The potential of harakeke fibre as reinforcement in
polymer matrix composites including modelling of long harakeke fibre
composite strength. Composites Part A 2015;76:44–53.
[15] Carr DJ, Cruthers NM, Laing RM, Niven BE. Fibers from three cultivars of New
Zealand flax (Phormium tenax). Text Res J 2005;75(2):93–8.
[16] Pickering K, Beckermann G, Alam S, Foreman N. Optimising industrial hemp
fibre for composites. Composites Part A 2007;38(2):461–8.
[17] Pickering K. Properties and performance of natural-fibre
composites. Cambridge, England: Woodhead Publishing; 2008.
[18] Cheng S, Lau K-t, Liu T, Zhao Y, Lam P-M, Yin Y. Mechanical and thermal
properties of chicken feather fiber/PLA green composites. Composites Part B
2009;40(7):650–4.
[19] Huson MG, Bedson JB, Phair NL, Turner PS. Intrinsic strength of wool fibres.
Asian-Australas J Anim Sci 2000;13:267.
[20] Gashti MP, Gashti MP. Effect of colloidal dispersion of clay on some properties
of wool fiber. J Dispersion Sci Technol 2013;34(6):853–8.
[21] Niu M, Liu X, Dai J, Hou W, Wei L, Xu B. Molecular structure and properties of
wool fiber surface-grafted with nano-antibacterial materials. Spectrochim
Acta Part A Mol Biomol Spectrosc 2012;86:289–93.
[22] Zhan M, Wool RP. Mechanical properties of chicken feather fibers. Polym
Compos 2011;32(6):937–44.
[23] Efendy MGA, Pickering KL. Comparison of harakeke with hemp fibre as a
potential reinforcement in composites. Composites Part A 2014;67:259–67.
[24] Cheung HY, Ho MP, Lau KT, Cardona F, Hui D. Natural fibre-reinforced
composites for bioengineering and environmental engineering applications.
Composites Part B 2009;40(7):655–63.
[25] Charlet K, Baley C, Morvan C, Jernot JP, Gomina M, Breard J. Characteristics of
Hermes flax fibres as a function of their location in the stem and properties of
the derived unidirectional composites. Composites Part A 2007;38
(8):1912–21.
[26] Li Y, Ma H, Shen Y, Li Q, Zheng Z. Effects of resin inside fiber lumen on the
mechanical properties of sisal fiber reinforced composites. Compos Sci
Technol 2015;108:32–40.
[27] Madsen B, Lilholt H. Physical and mechanical properties of unidirectional
plant fibre composites – an evaluation of the influence of porosity. Compos
Sci Technol 2003;63(9):1265–72.
[28] Beg MDH. The improvement of interfacial bonding, weathering and recycling
of wood fibre reinforced polypropylene composites. PhD thesis. Hamilton,
New Zealand: University of Waikato; 2007.
[29] Beckermann G. Performance of hemp-fibre reinforced polypropylene
composite materials. PhD thesis. Hamilton, New Zealand: University of
Waikato; 2007.
[30] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity and
stiffness. Compos Sci Technol 2009;69(7–8):1057–69.
[31] Madsen B, Thygesen A, Lilholt H. Plant fibre composites – porosity and
volumetric interaction. Compos Sci Technol 2007;67(7–8):1584–600.
[32] Hargitai H, Rácz I, Anandjiwala RD. Development of hemp fiber reinforced
polypropylene composites. J Thermoplast Compos Mater 2008;21(2):165–74.
[33] Shah DU, Schubel PJ, Licence P, Clifford MJ. Determining the minimum,
critical and maximum fibre content for twisted yarn reinforced plant fibre
composites. Compos Sci Technol 2012;72(15):1909–17.
[34] Matthews FL, Rawlings RD. Composite materials: engineering and
science. Cambridge, England: Woodhead Publishing; 1999.
[35] Vallejos ME, Espinach FX, Julian F, Torres L, Vilaseca F, Mutje P.
Micromechanics of hemp strands in polypropylene composites. Compos Sci
Technol 2012;72(10):1209–13.
[36] Andersons J, Sparnins E, Joffe R. Stiffness and strength of flax fiber/polymer
matrix composites. Polym Compos 2006;27(2):221–9.
[37] Huber T, Mussig J. Fibre matrix adhesion of natural fibres cotton, flax and
hemp in polymeric matrices analyzed with the single fibre fragmentation
test. Compos Interfaces 2008;15(2–3):335–49.
[38] Sun ZY, Han HS, Dai GC. Mechanical properties of injection-molded natural
fiber-reinforced polypropylene composites: formulation and compounding
processes. J Reinf Plast Compos 2010;29(5):637–50.
[39] Joffe R, Andersons J, Wallstrom L. Interfacial shear strength of flax fiber/
thermoset polymers estimated by fiber fragmentation tests. J Mater Sci
2005;40(9–10):2721–2.
[40] da Silva ILA, Bevitori AB, Lopes FPD, Monteiro SN. Pullout test of jute fiber to
evaluate the interface shear stress in polyester composites. In: Monteiro SN,
Verhulst DE, Anyalebechi PN, Pomykala JA, editors. EPD congress 2011.
Boston, MA: Wiley; 2011. p. 359–65.
[41] Nando GB, Gupta BR. Short fibre-thermoplastic elastomer composites. In: De
SK, White JR, editors. Short fibre-polymer composites. Cambridge,
England: Woodhead Publishing; 1996. p. 84–115.
[42] Sreekumar PA, Joseph K, Unnikrishnan G, Thomas S. A comparative study on
mechanical properties of sisal-leaf fibre-reinforced polyester composites
prepared by resin transfer and compression moulding techniques. Compos
Sci Technol 2007;67(3–4):453–61.
[43] Devi LU, Bhagawan SS, Thomas S. Mechanical properties of pineapple leaf
fiber-reinforced polyester composites. J Appl Polym Sci 1997;64(9):1739–48.
[44] Holbery J, Houston D. Natural-fiber-reinforced polymer composites in
automotive applications. JOM 2006;58(11):80–6.
[45] Summerscales J, Dissanayake NPJ, Virk AS, Hall W. A review of bast fibres and
their composites. Part 1 – fibres as reinforcements. Composites Part A
2010;41(10):1329–35.
[46] dos Santos PA, Giriolli JC, Amarasekera J, Moraes G. Natural fibers plastic
composites for automotive applications. In: 8th Annual automotive
composites conference and exhibition (ACCE 2008). Troy, MI: SPE
Automotive & Composites Division; 2008. p. 492–500.
[47] Faruk O, Bledzki AK, Fink HP, Sain M. Progress report on natural fiber
reinforced composites. Macromol Mater Eng 2014;299(1):9–26.
[48] Chen P, Lu C, Yu Q, Gao Y, Li J, Li X. Influence of fiber wettability on the
interfacial adhesion of continuous fiber-reinforced PPESK composite. J Appl
Polym Sci 2006;102(3):2544–51.
[49] Wu XF, Dzenis YA. Droplet on a fiber: geometrical shape and contact angle.
Acta Mech 2006;185(3–4):215–25.
[50] Bénard Q, Fois M, Grisel M. Roughness and fibre reinforcement effect onto
wettability of composite surfaces. Appl Surf Sci 2007;253(10):4753–8.
[51] Sinha E, Panigrahi S. Effect of plasma treatment on structure, wettability of
jute fiber and flexural strength of its composite. J Compos Mater 2009;43
(17):1791–802.
[52] Liu ZT, Sun C, Liu ZW, Lu J. Adjustable wettability of methyl methacrylate
modified ramie fiber. J Appl Polym Sci 2008;109(5):2888–94.
[53] Ragoubi M, Bienaimé D, Molina S, George B, Merlin A. Impact of corona
treated hemp fibres onto mechanical properties of polypropylene composites
made thereof. Ind Crops Prod 2010;31(2):344–9.
[54] Gassan J, Gutowski VS. Effects of corona discharge and UV treatment on the
properties of jute-fibre epoxy composites. Compos Sci Technol 2000;60
(15):2857–63.
[55] Seki Y, Sever K, Sarikanat M, Güleç HA, Tavman IH. The influence of oxygen
plasma treatment of jute fibers on mechanical properties of jute fiber
reinforced thermoplastic composites. In: 5th International advanced
technologies symposium (IATS’09), May 13–15, 2009, Karabük, Turkey;
2009. p. 1007–10.
[56] Cao Y, Sakamoto S, Goda K. Effects of heat and alkali treatments on
mechanical properties of kenaf fibers. Presented at 16th international
conference on composite materials, 8–13 July, 2007, Kyoto, Japan.
[57] Rong MZ, Zhang MQ, Liu Y, Yang GC, Zeng HM. The effect of fiber treatment on
the mechanical properties of unidirectional sisal-reinforced epoxy
composites. Compos Sci Technol 2001;61(10):1437–47.
[58] Huber T, Biedermann U, Muessig J. Enhancing the fibre matrix adhesion of
natural fibre reinforced polypropylene by electron radiation analyzed with
the single fibre fragmentation test. Compos Interfaces 2010;17(4):371–81.
[59] Beg MDH, Pickering KL. Mechanical performance of Kraft fibre reinforced
polypropylene composites: influence of fibre length, fibre beating and
hygrothermal ageing. Composites Part A 2008;39(11):1748–55.
[60] Singh B, Gupta M, Verma A. Influence of fiber surface treatment on the
properties of sisal-polyester composites. Polym Compos 1996;17(6):910–8.
[61] Beckermann GW, Pickering KL. Engineering and evaluation of hemp fibre
reinforced polypropylene composites: fibre treatment and matrix
modification. Composites Part A 2008;39(6):979–88.
[62] Kabir MM, Wang H, Lau KT, Cardona F. Chemical treatments on plant-based
natural fibre reinforced polymer composites: an overview. Composites Part B
2012;43(7):2883–92.
[63] Bera M, Alagirusamy R, Das A. A study on interfacial properties of jute-PP
composites. J Reinf Plast Compos 2010;29(20):3155–61.
[64] Gomes A, Matsuo T, Goda K, Ohgi J. Development and effect of alkali
treatment on tensile properties of curaua fiber green composites. Composites
Part A 2007;38(8):1811–20.
[65] Ibrahim NA, Hadithon KA. Effect of fiber treatment on mechanical properties
of kenaf fiber-ecoflex composites. J Reinf Plast Compos 2010;29(14):2192–8.
[66] Goda K, Sreekala M, Gomes A, Kaji T, Ohgi J. Improvement of plant based
natural fibers for toughening green composites—effect of load application
during mercerization of ramie fibers. Composites Part A 2006;37(12):
2213–20.
[67] Islam MS, Pickering KL, Foreman NJ. Influence of alkali treatment on the
interfacial and physico-mechanical properties of industrial hemp fibre
reinforced polylactic acid composites. Composites Part A 2010;41(5):
596–603.
[68] Kabir MM, Wang H, Lau KT, Cardona F, Aravinthan T. Mechanical properties of
chemically-treated hemp fibre reinforced sandwich composites. Composites
Part B 2011;43(2):159–69.
[69] Sawpan MA, Pickering KL, Fernyhough A. Improvement of mechanical
performance of industrial hemp fibre reinforced polylactide biocomposites.
Composites Part A 2011;42(3):310–9.
[70] Hill CAS, Khalil HPS, Hale MD. A study of the potential of acetylation to
improve the properties of plant fibres. Ind Crops Prod 1998;8(1):53–63.
[71] Bledzki AK, Mamun AA, Lucka M, Gutowsk VS. The effects of acetylation on
properties of flax fibre and its polypropylene composites. Express Polym Lett
2008;2(6):413–22.
110 K.L. Pickering et al. / Composites: Part A 83 (2016) 98–112