You are on page 1of 9

Industrial Crops and Products 53 (2014) 365–373

Contents lists available at ScienceDirect

Industrial Crops and Products

journal homepage:

Surface and thermal characterization of natural fibres treated

with enzymes
Michael George, Paolo G. Mussone, David C. Bressler ∗
Biorefining Conversions and Fermentations Laboratory, Department of Agricultural, Food and Nutritional Science, University of Alberta, T6E 2P5 Edmonton,

a r t i c l e i n f o a b s t r a c t

Article history: Natural fibres are a potential replacement for glass fibre in composite materials. Inherent advantages
Received 19 September 2013 such as low density, biodegradability and comparable specific mechanical properties (relative to glass
Received in revised form fibre composites) make natural fibres an attractive option. However, limitations such as poor thermal
25 November 2013
stability, moisture absorption and poor compatibility with polymeric matrices are challenges that need
Accepted 23 December 2013
to be resolved. The primary objective of this research was to study the effect of five enzymatic systems
on the surface chemical, morphological and thermal properties of natural fibres. Flax and hemp fibres
were treated with hemicellulases, pectinases and oxidoreductase. Surface and thermal properties were
Surface measured using X-ray photoelectron spectroscopy (XPS), thermal gravimetric analysis (TGA), scanning
Thermal characterization electron microscopy (SEM) and force tensiometry. Each treatment rendered the surface topography of
Natural fibres both fibres free of contaminants and exposed the individual fibre bundles. Treatment with hemicellulase
and pectinase improved the thermal properties for both fibres. XPS measurements confirmed reduction
of the hemicellulosic content of both fibres for xylanase and pectinases (polygalacturonase and pectin-
methylesterase). Removal of amorphous hemicellulosic material from the fibre surface and consequent
exposure of the crystalline cellulose network resulted in a lower contact angle for all the treated sam-
ples. This work demonstrated that enzymes offer an inexpensive and environmentally attractive option
to improve the surfaces of natural fibres for composite applications.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction studied and a comprehensive review was compiled by

Mukhopadhay and Fangueiro (2009). Methods such as steam
High-performance composite materials require a dispersed explosion (Kalia et al., 2009), plasma treatment (Yuan et al., 2004)
phase with elevated surface wettability and elevated cohesion with and corona discharge (Gassan and Gutowski, 2000) promote
the polymeric matrix in order to attain efficient transfer of stress separation of fibre bundles into more homogenous structures that
within the structure. While natural fibres constitute an attractive can be evenly deposited in one direction in the matrix materials.
alternative to glass fibres owing, their limited thermal stability An alternative to chemical and physical methods is repre-
and incompatibility with synthetic polymer matrices constitute a sented by the rapidly expanding use of biological agents such as
significant challenge toward industrial implementation of these fungi (Pickering et al., 2007; Gulati and Sain, 2006) and enzymes
materials platforms (Mohanty et al., 2002; Milanese et al., 2012; (Gustavsson et al., 2005; Kharazipour et al., 1997 and Pietak et al.,
Fei et al., 2009). Several stategies have been investigated to address 2007). Biological modifications offer several advantages over chem-
these deficiencies. For examples, mercerization (Mwaikambo and ical and physical methods. They can selectively remove hydrophilic
Ansell, 2002), acetylation (Bledzki et al., 2008) and silane treat- pectic and hemicellulosic material and require lower energy input.
ment (Anyakora and Abubakre, 2011) are known either to reduce Enzymatic systems, in addition, can be recycled after each use (Li
hydrogen bonding within the complex structure or by reacting and Pickering, 2008). Li and Pickering (2008) used chelators and
with the many hydroxyl groups on the surface (Gonzalez et al., enzymes to separate hemp fibres into individual bundles. They
1999; Tserki et al., 2005). Physical methods have also been found that the crystallinity (X-ray diffraction) and thermal prop-
erties (thermal gravimetric analysis) improved after separating the
bundles. In another study, Pietak et al. (2007) investigated the sur-
∗ Corresponding author. Tel.: +1 780 492 4986; fax: +1 780 492 4265. face wettability of natural fibres using atomic force microscopy and
E-mail addresses: Mngeorge@Ualberta.Ca (M. George), contact angle measurements. They measured an increase in adhe- (D.C. Bressler). sion force for treated samples because enzymatic and chemical

0926-6690/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
366 M. George et al. / Industrial Crops and Products 53 (2014) 365–373

treatments (steam explosion) removed the primary cell wall expos- Chemicals. Sulfuric acid (98%, mol wt. 98.075 g/mol) and calcium
ing the secondary wall. The same study provides clear evidence of carbonate (99%, mol wt. 100.09 g/mol) were also sourced from
hemicellulose removal from hemp fibres. Fisher Scientific. D-(+)-Glucose (99.5%, mol wt. 180.16 g/mol), D-
Enzymatic treatments have also been used to modify the phys- (+)-Galactose (99.5%, mol wt. 180.16 g/mol), D-(+)-xylose (99%, mol
ical network of natural fibres. Kardas et al. (2009) investigated the wt. 150.1 g/mol) were obtained from Sigma. Distilled water was
effect of four enzyme treatments (Lipase A, Lipase AK, Lipozyme and used for all analyses.
Esterase) on the micro-topography of polyester fabric. They found
that the esterase preparation was the most effective method for 2.2. Washing of natural fibres
producing a more uniform and homogenous texture of the fabric
material. Saleem et al. (2007) treated bast fibres with pectinase and Fibres were washed using 2% (w/v) Sparkleen (No. 1) industrial
then reported the mechanical characteristics of the reinforced ther- detergent for 1 h at 70 ◦ C to ensure removal of dirt and any surface
moplastic composites. According to their study, hemp fibres treated contaminant that may affect the enzymes activity. Samples were
with 8% of the enzyme in maleic anhydride medium, were charac- filtered and washed several times with distilled water. Fibres were
terized by increased tensile strength, flexural strength and moduli dried in a convection oven at 80 ◦ C for 5 h and stored in a desiccator
of elasticity. These improvements were attributed to higher aspect (Bismarck et al., 2001).
ratio of the fibre bundles which resulted in better dispersion within
the polymeric matrix (polypropylene). A recent studied conducted 2.3. Natural fibre composition
on bamboo fibres using a number of enzymes (xylanase, cellulase,
pectin lyase and laccase) revealed that the different systems were The methods outlined by Ramadevi et al. (2012) were adopted to
effective in improving fibre fineness (Liu et al., 2012), presumably estimate the lignin, hemicellulose and cellulose content. Briefly, the
as a result of the removal of the more polar hemicellulosic fraction. lignin content was measured by placing 2 g of sample in a 500 ml
The scope of this study was to investigate the effect of beaker and 15 ml of 72% sulfuric acid was added and agitated at
five commercially available enzyme systems (xylanase, pectin- 100 rpm for three hours at 25 ◦ C. 200 ml of water was then added
methylesterase, polygalacturonase, laccase and a xylanase with to the mixture and boiled for two hours. Once cooled, after 24 h,
cellulase background) on the surface of two different sources of bast lignin was transferred to a crucible and washed with hot water.
fibres, hemp and flax. It should be noted that each of these enzyme The crucible with its content was dried at 105 ◦ C and weighed every
system may have accessory enzymes such ␣-glycosidase and arabi- hour until a constant weight was reached.
nosidase. Fundamental surface chemistry and thermal information The holocellulose content was measured by placing 160 ml of
collected in this work provided the basis for better understand- water, 0.5 ml of acetic acid, 1.5 g of sodium chloride and 3 g of fibre
ing the properties of natural fibres and how these may influence in a beaker. The beaker was placed in a water bath maintained at
the fibre-matrix interface. Enzymes enable the modification of the 75 ◦ C for an hour, after which 0.5 ml of acetic acid was added with
fibre surface properties with limited or no effect on the bulk prop- 2.5 g of sodium chloride. This was repeated for another two hours.
erties because, owing their size percolation into the inner structure Once completed, the beaker was then placed in an ice bath and
is impeded. For example, Boisset et al. (2001) while studying the cooled. The isolated holocellulose was sequentially washed with
action of a recombinant cellulase on the cross section of cottons acetone, ethanol and water. The sample was finally dried at 105 ◦ C
fibre for applications in the textile industry found that there was until constant weight was reached.
no indication of enzyme penetration or damage to the interior of The content of ␣-cellulose was measured by adding 10 ml of
the fibres. 18% (w/v) sodium hydroxide to 2 g of the isolated holocellulose.
To the best of our knowledge, this study is one of the first The fibres were continuously agitated at 120 rpm and kept at 20 ◦ C.
that examines the influence of solely enzymes in the absence of Every 5 min, 10 ml of the sodium hydroxide solution was added for
any mediator (chemicals) on thermal properties of natural fibres another half hour. 35 ml of water was added to the beaker and kept
specifically for composites applications. Removal of pectic and for another hour. The holocellulose residue was isolated by washing
hemicellulosic materials resulted in improved thermal proper- with 100 ml of 8.5% sodium hydroxide, 200 ml of water, 15 ml of 10%
ties based on thermal gravimetric analysis. X-ray photoelectron (w/v) acetic acid and water. The contents of the crucible were dried
spectroscopy and contact angles indicated the removal of the at 105 ◦ C until constant weight was reached.
hygroscopic primary cell wall. The findings reported serves to link The hemicellulose content was calculated by subtracting the
enzymes as a pretreatment method for natural fibres to be used in hollocelulose from the ␣-cellulose content (Kalia et al., 2009).
composite applications. ASTM Method D4442-07 was followed to determine the mois-
ture content of the flax and hemp samples. Approximately 0.5 g
(done in triplicate) of sample was weighed in a pre-weighed
2. Experimental and dried tin pan. The pan was placed in a convection oven at
105 ◦ C ± 5 ◦ C for 5 h. The sample pan was removed, cooled in a
2.1. Materials desiccator and then weighed. The sample was heated, cooled and
re-weighed until constant mass. The moisture content was calcu-
Mechanically processed hemp and flax samples were pro- lated based on percentage mass lost.
vided by the Alberta Biomaterials Development Centre located ASTM Method E1755-01 was used to estimate the ash con-
in Vegreville, Alberta. The samples were placed in air tight bags tent of the different fibre samples. The method involved weighting
and stored at 4 ◦ C. All enzymes were provided by Novozymes approximately 0.5 g (done in triplicate) of fibre sample into a pre-
(Bagsvaerd, Denmark) and stored at 4 ◦ C. Sodium acetate (99%, mol weighed crucible. The crucibles were heated to 575 ◦ C ± 25 ◦ C for
wt. 82.03 g/mol), glacial acetic acid (99.7%, mol wt. 60.05 g/mol), 3 h or until all the carbon was eliminated. To avoid causing flames,
sodium phosphate dibasic (99%, mol wt. 141.96 g/mol) and sodium the sample was heated to 250 ◦ C and gradually increased to 575 ◦ C.
hydroxide (99%, mol wt. 40.00 g/mol) were obtained from Fisher The residue left after 3 h was cooled in a desiccator and weighed.
Scientific. Sodium citrate monohydrate (99%, mol wt. 214.11 g/mol) The contents were heated in a convection oven at 105 ◦ C until con-
was purchased from Sigma-Aldrich. Sodium phosphate monoba- stant weight. The ash content was calculated as a mass percentage.
sic (99%, mol wt. 119.98 g/mol) was obtained from Acros Organics. The Dumas method or Combustion Nitrogen Analysis (CNA)
Citric acid (99%, mol wt. 192.13 g/mol) was sourced from EMD method was used to determine the total organic and inorganic
M. George et al. / Industrial Crops and Products 53 (2014) 365–373 367

Table 1 and O1s (530 eV) (Sgriccia et al., 2008). To maximize the analysis
Characteristics of each enzyme system with corresponding activity at optimum
incident area, a spot size of 30 ␮m of the fibre surface was used
conditions. For each system, pH was maintained using buffers.
to record each spectrum. Survey spectra were acquired using pass
Enzyme Optimum conditions Activity energy of 50 eV for low resolution and 20 eV for high resolution.
pH Temperature (◦ C) The fibres were mounted on aluminum foil and then compressed
using nickel plates. The sample assembly was placed in the sample
Xylanase 7 70 1000 AXU/g
Xylanase (10% cellulase) 6 50 2500 FXU-S/g processing chamber of the spectrometer for analysis. An eight (8)
Polygalacturonase 4 45 3800 PGNU/ml channel multi-detector was employed for analysis (Zafeiropoulos
Laccase 7 50 1000 LAMU/g et al., 2003).
Pectinmethylesterase 5 45 5 PEU/ml Data treatment was performed using CasaXPS program (Casa
Software Ltd., UK). Mole fractions were calculated using normalized
peaks based on acquisition parameters after a linear background
subtraction and consideration of experimental sensitivity factors.
nitrogen content of the different samples. A LECO TruSpec CN
C1s spectra were analyzed with a Gaussian product function, by
Instrument was used for all analyses. The nitrogen content of the
maintaining the FWHMs of all components to within the range of
samples was used to estimate the protein concentration with a con-
1.200–1.600 (Zafeiropoulos et al., 2003).
stant factor approved by the Canadian Grain Commission (Williams
et al., 1998). In all cases, approximately 0.1 g (done in triplicate)
finely chopped fibre was used. The protein content was estimated
2.5.3. Force tensiometry/contact angle measurement
by multiplying a protein factor of 6.25 by the total nitrogen content
A 700 Sigma One Attension tensiometer (Biolin Scientific)
(Williams et al., 1998).
was used to measure the contact angle. All samples were pre-
conditioned at 80 ◦ C for 3 h prior to analysis. The Washburn
2.4. Enzymatic treatment capillary rise method was used to characterize the fibre bundles
(Aranberri et al., 2003; Baley et al., 2006). For each experiment,
Approximately 1.0 gram of fibre was weighed into a 125 ml triplicate analysis was done for each sample with three cycles
Erlenmeyer flask. Optimal conditions for each enzymatic reaction of immersion (three measurements taken on a single fibre). The
were provided by Novozymes (Table 1). Concentrations of 2, 6 and first 0.05 mm of each fibre was ignored because the edges of the
10% (% w/v) enzyme stock with corresponding enzyme activity fibres are not spherical and can influence the result. An immer-
as specified in Table 1 were used to treat each fibre type. For all sion speed of 5 mm/min was used for both advancing and receding
experiments, the liquid (ml) to fibre (g) ratio was maintained at measurements. Given the nature of the technique, both advanc-
50:1 to facilitate complete wetting of fibres. Enzymatic treatments ing (penetration of the fibre into liquid) and receding (recession of
were conducted for 90 min under constant agitation (80 rpm) at the fibre from liquid) contact angles were calculated for each cycle.
the optimum temperature in a standard water bath. Enzymes were Equation 1 was used to estimate the contact angle for each sample.
deactivated by heating at 90 ◦ C for 10 min. Fibres were washed with
excess warm water to remove traces of enzyme and buffer reagents.
All samples were dried at 80 ◦ C for 5 h and stored in polyethylene
Wetting force = Liquid vapor surface tension ∗ perimeter fibre (P)
bags for subsequent analysis. All experiments were carried out in
triplicate. ∗cos (1)

2.5. Characterization of fibres where  is the unknown contact angle. The contact angle for each
sample was calculated using the measured wetting force.
Prior to all analyses, samples were pre-conditioned at 80 ◦ C
for 5 h to maintain constant moisture content for all treated and
untreated samples. 2.5.4. Thermogravimetric analysis (TGA)
Experiments were conducted using a Thermal Analysis Instru-
2.5.1. Morphological characterization ments TGA Q50 (TA Instruments) apparatus under a flow of
Micrographs of fibre surfaces of untreated and enzyme treated nitrogen to study the effects of heating on stability of the different
fibres were taken using Hitachi S-2700 Scanning Electron Micro- treated natural fibres. Platinum pans were used given the high tem-
scope (SEM) equipped with a Princeton Gamma Tech (PGT) IMIX peratures and the ease of cleaning. The temperature range selected
digital imaging system and a Prism Intrinsic germanium (IG) was from room temperature to 600 ◦ C at a rate of 10 ◦ C per minute.
detector. A gold putter coater was used to induce conductivity For each sample, triplicate runs were done. All results were repro-
for all samples. A resolution of 4 nm was used for all samples. duced to 5% error or better.
0.1 ± 0.005 g of fibre was mounted on conductive adhesive tape,
sputtered coated with gold palladium and observed using a voltage
of 15–20 kV (Reddy and Yang, 2005). 2.6. Statistical analysis

2.5.2. X-ray photoelectron spectroscopy (XPS) All experiments were replicated at least three times and results
XPS was used to study the elemental composition and the O/C were expressed as mean value ± standard deviation. The statistical
ratio of the surface of the different fibres following treatment. analyses of the data were conducted using the statistical software
Samples were subjected to high and low resolution scans. Approx- package SAS Version 9.4. It is important to note, that each system
imately 0.2 ± 0.005 g of fibre was pressed into a disc with diameter was compared to the control and not among each other given the
of about 0.5 mm for analysis. A Kratos Ultra 165 X-ray Photoelec- difference in activity for the enzymes and the different mode of
tron Spectrometer with energy spectra ranging from 0 to 1000 eV substrate attack. To identify significant differences between mean
was used to study all samples. O/C ratios for the different sam- values for control and a given enzyme system, Kruskal Wallis Test
ples were compared to the 0.83 for cellulose, hemicellulose, and was applied to the data populations involved, with a 95% confidence
pectin and 0.35 for lignin. Samples were scanned for C1s (284 eV) level (P < 0.05).
368 M. George et al. / Industrial Crops and Products 53 (2014) 365–373

3. Results and discussion It is plausible that the removal of lignin exposes the bundle as
seen in both hemp and flax systems. On the other hand, a combina-
3.1. Characterization of flax and hemp tion of xylanase and cellulase (c) and polygalacturonase (d) treated
samples exhibited an increase in surface roughness and bundle
The chemical analysis of the major components of flax and hemp exposure (Karaduman et al., 2013). Xylanase is known to degrade
is presented in Table 2. With the exception of the cellulose content hemicellulosic based material which is found around the cellulosic
(measured at 74 and 81% on a dry weight basis for hemp and flax, fibre bundles as pockets of different branched and linear polysac-
respectively) the data collected was in line with the values reported charides (Prade, 1996). Xylanase attacks these pockets of polymers
by Mwaikambo and Ansell (2002). breaking ␤-1,4 bonds found in xylan releasing monomers result-
In another study, Tserki et al. (2005) found that the relative ing in the formation of cracks on the fibre surfaces. Hemp samples
amount of hemicellulose to lignin influences the kinetics of chem- treated with polygalacturonase showed clear evidence of removal
ical modification of wood, flax and hemp fibres. Interestingly, their of material from the fibre surface possibly as a result of the degra-
characterization showed 7 and 15% (weight%) for minerals and dation of polygalacturonan, a major component of pectin found
proteins, respectively. This fraction constitutes a large portion of predominantly in the primary cell wall (Bateman and Basham,
dissolved and colloidal matter which is known to affect the activ- 1976). Baracat et al. (1989) showed that this family of enzymes
ity of enzymes and fungi. The moisture content for the samples disrupts the pectic middle lamella and primary cell walls exposing
studied was comparable with the results of Tserki et al. (2005) of the cellulosic network during the degumming of fruits for food and
approximately 9.1% (weight %). The experimental samples had a fermentation applications. Using this base of knowledge, one can
high amount of protein when compared to the values reported remove the hygroscopic pectin content of natural fibres and this
by Bismarck et al. (2001) in Table 2. The values available in the should translate into an improvement in thermal resistance. The
literature do not specify the nature (organic or otherwise) of the work reported for the hemp samples are in agreement with the lit-
nitrogen source. The test utilized in this study (DUMAS combus- erature. Reddy and Yang (2005) suggested in a review that the use
tion method) estimates the total nitrogen instead of solely organic of enzymes such as pectinases and hemicellulases for dry retting
nitrogen. Apart from low cellulose content, the composition of the will be an important area of focus in the coming future. Although
experimental fibres was comparable to literature values. the ultimate aim during retting and enhancing of fibres is similar,
Characterization of the treated samples based on the three the dosage, time and type of enzyme will impart different features.
main components namely cellulose, hemicellulose and lignin are In fact, when enhancing the fibres, the aim is to modify the surface
presented in Table 3. This set of data, integrated with XPS and con- only by selective removal of components but for retting the aim is
tact angle measurements enables a correlation between surface to remove the components that hold the network together produc-
composition and polarity. Xylanase + cellulase treated hemp were ing fibrils that can be modified. In retting, the enzymes removed the
characterized by a significant (p < 0.05) reduction in the cellulosic encrusting substances such as pectin and hemicellulose resulting
content while hemp treated with pectinases (polygalacturonase in fibres with greater surface roughness. It was their opinion that
and pectinmethylesterase) showed reduction in the hemicellulosic these fibres can then be treated to render them more hydrophobic
content. One plausible reason for the latter observation may be the so as to improve the adhesion with polymeric matrices.
removal of pectic matter (minor component ≈ 0.5–2% dry weight) Flax fibres treated with xylanase + cellulase showed the same
bonded to hemicellulosic content which leads to disintegration and pattern of surface roughness as did the hemp sample. The fibre
rupture of the network. The same trend was observed for these bundles were more exposed in the case of flax samples, possibly as
enzymes when applied to flax fibres. In addition, laccase treated flax a result of an initially higher amount of hemicellulose (Table 2) and
fibres were characterized with a significant reduction (p < 0.05) in less cellulose. Flax fibres treated with were also characterized by
the lignin content. The flax fibres were more susceptible to enzyme bundle fibrillation and increased in surface roughness. In summary,
attack based on the characterization of the treated fibres” it is evident from the SEM micrographs that enzyme treatment of
hemp and flax led to individual bundle exposure and a clean surface.
3.2. Morphological characterization
3.3. X-ray photoelectron spectroscopy
Five enzyme systems were used to treat hemp and flax samples.
The fibre sections were randomly selected. The possible influence of Most of the characterization studies of natural fibres using XPS
changes in stem morphology (fibres from different sections of stem) have been carried out on chemically treated samples (Zafeiropoulos
was eliminated via processing by standard methods approved et al., 2003; Williams et al., 2011) and Brigida et al., 2010). In addi-
by the Pulp and Paper Technical Association of Canada and in tion to modifying the surface of the fibres, chemical pretreatments
accordance with industrial standard methods in use at Alberta may also alter the inner structure adversely affecting the mechan-
Innovates Technology Futures (Vegreville, Alberta). The morpho- ical properties (Jayabal et al., 2012). On the other hand, enzymatic
logical changes for the different treated samples showed similar reactions are limited to the surface of the fibres because access to
pattern as represented in Fig. 1. In Fig. 1, the micrographs for an the inner structure is precluded. An example of the high resolution
untreated (a), laccase (b), xylanase + cellulase (c) and polygalac- carbon C1s spectra for hemp fibres showing the different fractions
turonase (d) treated samples. It is clear from the images when associated with each carbon is shown in Fig. 2. The parent peak was
compared to the control (untreated), the fibre bundles become partitioned into its different carbon signatures, depending on the
more visible. Gutierrez et al. (2012) demonstrated that laccase binding energy. The three major signatures studied were CH2 –CH2
in the presence of mediators (HBT) removed lignin, including due to carbon backbone, C–O and O–C–O at 285, 286.2–286.6 and
aliphatic derivatives from wood feedstocks. The aim of our work 287.3–288.1 eV, respectively.
was to use the enzyme solely (without any chemicals) and study Table 4 summarizes the data obtained using XPS and shows the
whether there were any surface changes due to the removal or changes in the atomic percentage of carbon involved in different
rearrangement of the lignin fragments. Researchers have studied bonds and the O/C ratio for each treatment. The data was obtained
the macromolecular arrangement of the different components of by fitting the carbon peak at 285 eV using Gaussian/Lorentzian
wood based and non wood based feedstocks. The widely accepted functions. Analysis of the data was done by measuring the areas
theory is that lignin forms a random wrap-like structure around under the curves (O and C) and computing the O/C ratios for
the cellulosic bundles (Mohanty et al., 2002). each enzyme system. The C peaks were partitioned into its
M. George et al. / Industrial Crops and Products 53 (2014) 365–373 369

Table 2
Chemical compositions for flax and hemp samples analyzed relative to values from the literature (given in mass percentage).

Component Experimental samples Literature valuesa

Hemp Flax Hemp Flax

Moisture (%) 7.34 ± 0.19 8.12 ± 0.18 10 8

Protein (% w/w) 6.34 ± 0.30 6.18 ± 0.28 2.1 3
Ash (% w/w) 2.31 ± 0.13 1.53 ± 0.19 – –
Cellulose 67.77 ± 2.25 64.08 ± 3.09 67 65
Lignin 7.24 ± 0.64 4.14 ± 0.42 3.3 2.5
Hemicellulose 12.14 ± 3.51 16.10 ± 3.02 16 16
Bismarck et al. (2001).

Table 3
Chemical compositions of the treated samples with respect to cellulose, hemicellulose and lignin.

Fibre System Component (% of dry weight)

Cellulose Hemicellulose Lignin

Hemp Untreated 67.77 ± 2.25 12.14 ± 3.51 7.24 ± 0.64

Xylanase 67.18 ± 1.46 7.89 ± 0.81 7.04 ± 0.61
Xylanase + cellulase 57.42 ± 2.12 11.80 ± 1.02 6.48 ± 1.35
Laccase 68.01 ± 2.48 10.97 ± 0.86 6.88 ± 0.05
Polygalacturonase 68.63 ± 4.58 6.40 ± 2.21 5.09 ± 1.10
Pectinmethylesterase 68.35 ± 1.31 7.45 ± 1.81 6.14 ± 1.17

Flax Untreated 64.08 ± 3.09 16.10 ± 3.02 4.14 ± 0.42

Xylanase 63.56 ± 2.65 11.37 ± 1.25 3.53 ± 0.09
Xylanase + cellulase 54.50 ± 1.31 8.69 ± 1.96 4.18 ± 0.89
Laccase 63.06 ± 2.09 13.35 ± 2.64 1.45 ± 0.12
Polygalacturonase 61.31 ± 1.99 2.55 ± 0.84 1.43 ± 0.28
Pectinmethylesterase 61.12 ± 1.72 4.15 ± 1.82 2.81 ± 0.06

Fig. 1. Micrographs of treated and untreated hemp; (a) control, (b) laccase, (c) xylanase + cellulase and (d) polygalacturonase. The red insets show the defibrillation in the
laccase and xylanase + cellulase treatments and the individual bundle exposure in the hemp treated with polygalacturonase. (For interpretation of the references to colour
in this figure legend, the reader is referred to the web version of the article.)
370 M. George et al. / Industrial Crops and Products 53 (2014) 365–373

Table 4
Abundance of the different C1s signatures for the different enzyme treated fibre samples and the corresponding O/C ratio.

Sample Treatment Signature (eV)-(Atomic%) O/C

CH2 –CH2 (285) C–O (286.2–286.7) O–C–O (287.3–288.1)

Hemp Control 35.41 ± 4.62 42.88 ± 1.34 20.08 ± 7.71 0.63 ± 0.09a
Xylanase + cellulase 36.76 ± 2.24 37.65 ± 2.92 25.22 ± 4.70 0.52 ± 0.04a
Xylanase 25.98 ± 1.26 41.50 ± 9.16 18.38 ± 2.82 0.65 ± 0.02a
Polygalacturonase 24.93 ± 2.70 54.99 ± 2.31 20.08 ± 2.23 0.71 ± 0.05a
Pectinmethylesterase 32.074 ± 5.89 37.88 ± 2.02 29.38 ± 6.17 0.65 ± 0.06a
Laccase 25.03 ± 3.87 49.80 ± 5.34 22.03 ± 3.08 0.45 ± 0.09b

Flax Control 24.55 ± 0.60 54.57 ± 0.89 20.89 ± 1.49 0.67 ± 0.01a
Xylanase + cellulase 27.97 ± 0.94 47.88 ± 2.88 22.16 ± 0.98 0.52 ± 0.04c
Xylanase 29.31 ± 0.69 49.43 ± 2.53 21.26 ± 2.80 0.67 ± 0.03a
Polygalacturonase 46.21 ± 5.64 38.37 ± 4.32 15.42 ± 1.61 0.18 ± 0.01d
Pectinmethylesterase 43.36 ± 1.96 40.45 ± 2.09 16.19 ± 0.80 0.18 ± 0.00b
Laccase 23.65 ± 1.84 52.68 ± 2.57 23.65 ± 2.65 0.35 ± 0.16e

(a,b,c,d,e) Treatments with the different letters are statistically different at a confidence level of 0.05 (n = 3) when compared to the control (not among enzymes). Kruskal
Wallis Non-Parametric statistical analysis was used to analyze all samples. This test was used given the small sample size and the non-parametric pattern of the data.

signatures using CasaXPS software. These components correspond (Bateman and Basham, 1976). These constituents are found pri-
to CH2 –CH2 , C–O and O–C–O as outlined by Zafeiropoulos et al. marily in the cell wall and degradation exposes aromatic rich lignin
(2003). and its derivatives which cover the cellulose, hence the decrease in
The abundance of the CH2 CH2 bonds, which are attributed to O/C. On the other hand, a combination of xylanase and 10% cel-
waxy substances and carbon backbone (Jayabal et al., 2012), was lulase decreased the C–O while increasing the C–OH functionality
reduced after every enzymatic treatment in the case of hemp. It indicating that the hemicellulosic fraction was degraded exposing
should be noted that all enzyme treatments were compared to a the more crystalline cellulose. Pectinmethylesterase and xylanase
control and not among treatments given that each enzyme sys- in the presence of cellulase background had no significant effect
tem is distinct both in the relative amounts used and the mode of (p < 0.05) on the surface chemical composition when compared to
action. The distribution of heterogeneous layers of waxes on the the control experiment.
fibre surfaces was also observed by the SEM micrographs, in good Flax treated with xylanase and laccase showed no difference
agreement with literature results (Zafeiropoulos et al., 2003). There from the control for the surface chemical composition and will
was no significant (p < 0.05) change in the C–O signature for most not be discussed further. However, flax treated with polygalactu-
treatments. ronase, an enzyme highly specific in attacking pectin derivates in
The primary component of hemp and flax is cellulose while the primary cell wall and middle lamella of the fibres (Bateman
hemicellulose amounts to approximately 15% hemicellulose. and Basham, 1976), exhibited a significant reduction in the O/C
According to Zafeiropoulos et al. (2003), these two major compo- ratio. Bateman and Basham (1976) did a comprehensive review
nents with a range of monomers have an O/C ≈ 0.83. For the hemp on microbial enzymes capable of degrading plant cell wall com-
treated samples, when compared to the control, polygalacturonase ponents of different material and found that polygalacturonase
treated fibres had an O/C of ≈0.71, suggesting that the degradation was one of the most effective when it comes to pectin degrada-
of the middle lamella and primary cell wall may have exposed the tion agents during fruit ripening. A plausible explanation for the
less hygroscopic cellulosic network. This is also supported by the reduction of the O/C can be exposure of lignin components after
increase in the C–O functionality. In the case of hemp treated with the polygalacturonan removal. This observation is also supported
xylanase, there was a decrease in C–OH functionality, possibly as by the reduction in C–OH signature of hemicellulose on the fibre
a result of hemicellulose removal which left the cellulosic bundles surfaces and a corresponding increase in aliphatic (CH2 –CH2 ) sig-
covered with lignin. No significant change in the O/C ratio was mea- nature (due to waxes and derivatives of lignin). The flax system
sured, indicating that the cellulose structure probably remained was more susceptible to the enzyme treatments used. Both hemi-
intact (Boisset et al., 1997). cellulases (pectinmethylesterase and polygalacturonase) reduced
Pectinmethylesterase decreased the O/C ratio of flax sample as the O/C ratio as a result of hemicellulose removal and exposure
a result of pectin de-esterification into pectic acid and methanol of lignin on the surface. The O/C ratio for lignin was 0.35 while
for hemicellulose material it was approximately 0.85, hence the
removal of the latter will result in concentration of the former on
the surface, reducing the O/C ratio. The geometrical orientation of
the individual fibres in a bundle provides a plausible explanation
for the reactivity differences of these fibres (Wiener et al., 2003).
A smaller lumen and 7-sided nature of the fibre cross section with
sharp peaks makes flax fibres more susceptible to enzymatic attack.
The folds in the cross section allow greater surface access for the
enzymes, resulting in greater hemicellulose removal (Wiener et al.,
2003). In summary, the activity of the different enzymes depends
on geometrical orientation of the individual fibres within a bundle
and chemical composition of the natural fibres.

3.4. Force tensiometry

Fig. 2. High resolution deconvoluted Carbon C1 XPS spectra for untreated hemp Table 5 shows the effect of each enzymatic treatment on the
fibres illustrating the different carbon abundance. contact angle of the fibres studied. The advancing contact angle
M. George et al. / Industrial Crops and Products 53 (2014) 365–373 371

Table 5 conformation of the individual fibres in a bundle while hemp fibres

Contact angle for the different enzyme treated hemp and flax.
are characterized by a polygonal shape with rounded peaks. This
System Enzyme Contact angle (◦ ) may be the reason for the greater susceptibility of the flax fibres to
enzymes because the structure allows greater accessibility for the
Control Sample
enzymes (Wiener et al., 2003).
Hemp Xylanase + 10% Cellulasea 89.13 ± 0.07 88.71 ± 0.03
The prevailing view, based on studies conducted using model
Xylanasea 89.13 ± 0.07 88.68 ± 0.04
Polygalacturonasea 89.06 ± 0.16 88.52 ± 0.03 systems such as hexenuronic acids (Aracri and Vidal, 2011), sug-
Pectinmethylesterase 89.06 ± 0.16 88.81 ± 0.17 ars (Prade, 1996) and fruit cell walls (Palanivelu, 2006) is that the
Laccasea 89.06 ± 0.02 88.70 ± 0.08 rate of degradation of these components depended on the compo-
Flax Xylanasea 88.97 ± 0.18 88.49 ± 0.04 sition of the material. The data presented in this research clearly
Xylanase + 10% Cellulasea 88.99 ± 0.04 88.59 ± 0.02 demonstrate that the modification of natural fibres using enzymes
Polygalacturonase 88.75 ± 0.13 88.80 ± 0.22 depends both on the chemical composition and geometrical orien-
Pectinmethylesterase 88.91 ± 0.17 88.64 ± 0.29
tation of the individual fibres within a bundle.
Laccase 88.80 ± 0.43 88.85 ± 0.20
Flax fibres are characterized by many contours (roughness)
Treated samples that were significantly different from the control (p < 0.05). within each bundle and hence a larger surface area compared to
the hemp fibres. The size of the enzymes is at least four orders of
for each system was presented because there was no difference magnitude smaller than that of an individual bundle and at least ten
between the advancing and receding contact angles (p < 0.05). Typ- times smaller than a fibril. It is reasonable to expect unencumbered
ically, differences between the advancing and receding angles may access to the crevices present on the flax fibre surface, resulting in
be due to geometric heterogeneity or small scale roughness and greater reactivity. Based on this observation, flax fibres may be con-
chemical heterogeneity. Despite the known heterogeneity of the sidered a better option for enzymatic treatments to produce natural
natural fibres, the absence of hysteresis in almost all cases indi- fibres suitable for composite materials.
cated the enzymatic treatment did remove the specific components
resulting in a clean fibre.
All enzyme treatments of hemp fibres resulted in significant 3.5. Thermogravimetric analysis
(p < 0.05) reductions of contact angle which are correlated to a
greater surface hydrophilicity (Pickering et al., 2007). It must be Fig. 3 shows a typical set of thermographs expressed as the
emphasized that the structural arrangement of the different chem- derivative weight lost function for xylanase + cellulase treated
ical component of a single fibre strand is complex. According to hemp. Two regions of significance are observed. The first region
Biagiotti et al. (2004) the fibre network consists of basically four of degradation (200–265 ◦ C) corresponds to hemicellulose degra-
distinct layers namely the secondary wall, primary cell wall, mid- dation and the second region (265–400 ◦ C) corresponds to cellulose
dle lamella and the layers (S1, S2) in between the primary and degradation (Yang et al., 2007). Tajvidi and Takemura (2008)
secondary walls. The outermost layer consists primarily of hemi- reported that lignin is the most thermally resistant, cellulose being
cellulose covered cellulose. Each section of the network because the intermediate and hemicellulose the least stable.
of the varying chemical components that are abundant, have their Results for the different enzymatic treatments are represented
unique characteristic polarity (Biagiotti et al., 2004). in Table 6. The parameters presented included the percentage
Specifically, xylanase in the absence of cellulase has been used in degradation (% weight), which corresponds to the weight lost from
the pulp and paper industry to facilitate lignin extraction and degra- the onset degradation temperature (≈260 ◦ C) to 400 ◦ C. The max
dation of xylan, a major fraction of the hemicellulose (Katapodis degradation temperature is the maximum of the derivative weight
et al., 2003). The removal of the xylan is known to expose the curve as illustrated in Fig. 3.
cellulose hydroxyl and carboxyl groups that were otherwise cov- Hemp fibres treated with laccase were characterized by a
ered. Similarly, xylanase treatment in the presence of the cellulase decrease in percentage degradation while no changes in the onset
background removed the hemicellulose content from the primary of degradation were observed. The maximum degradation temper-
cell walls of the fibre bundles and causes some degradation of the ature was lowered because the removal of lignin bonded cellulose.
cellulosic network as evident in the SEM micrograph. Polygalactu- Laccase is responsible for removing the lignin fraction of the fibres,
ronase was used to depolymerize pectin/pectic acid found in the thereby exposing the less stable hemicellulose and cellulose. This
middle lamellae by cleaving the glycosidic bonds via hydrolytic explains the increase in degradation of the flax treated samples, but
reactions. This leads to either crystalline cellulose or other hemicel- was contrary to the hemp treated samples for the same enzyme.
lulosic content within the primary wall (Palanivelu, 2006). Laccase This may be possible because hemp has a greater amount of
treated hemp resulted in a decrease in contact angle because the cellulose and less hemicellulose than the flax fibre. Removal of
removal of lignin exposes the cellulosic backbone (as supported by the lignin exposed the cellulosic network which is more resis-
increase in [C–OH] from XPS study) (Aracri and Vidal, 2011). Pectin- tant than hemicellulose. Hemp treated with pectinmethylesterase
methylesterase had no effect on the surface polarity for the hemp (PME), xylanase and xylanase + cellulase exhibited an improve-
fibres. The enzyme systems studied resulted in an increased sur- ment in thermal degradation. This is expected given that these
face hydrophilicity plausibly because of the removal of lignin and enzymes removed the hemicellulosic fraction of the fibre. Specif-
hemicellulose, exposing the cellulosic backbone. ically, PME removed the pectic content while xylanase degraded
Laccase and the pectinases had no significant effect (P < 0.05) the xylan backbone of the hemicellulose structure. There was a
on the flax fibres when compared to the control. However, the decrease in the maximum degradation temperature for the hemp
flax system when treated with both hemicellulases showed sim- samples when compared to the control because the components
ilar values when compared to the hemp samples. Interestingly, the removed resulted in a weaker network because off disruptions in
contact angles decreased more than when compared to the hemp bonding within the network. Hemp fibres treated with polygalac-
fibres treated with the same enzymes. As mentioned before, one turonase were characterized by lower thermal stability as indicated
plausible reason for this may be the geometrical orientation differ- by the increase in percentage degradation. This can be attributed to
ences between the two fibre types. Wiener et al. (2003) identified a the removal of polygalacturonan which weakens the primary cell
simple and robust method to differentiate between flax and hemp. wall, ultimately leading to a weaker macrostructure (Bateman and
Flax fibres have a larger surface area because of the 7-sided peak Basham, 1976).
372 M. George et al. / Industrial Crops and Products 53 (2014) 365–373

Fig. 3. Derivative weight lost of xylanase + cellulase treated hemp at three different concentrations depicting the different paramters measured.

Table 6
Thermal properties for different treated samples (n = 3, p < 0.05).

Fibre System Parameter

%Degradationa (%wt.) Onset degradationb (◦ C) Max. degradation temperaturec (◦ C)

Hemp Control 62.38 ± 0.08 275.13 ± 0.16 354.72 ± 0.52

Xylanase + cellulase 58.30 ± 0.64 277.08 ± 0.52 351.38 ± 0.31
Xylanase 54.81 ± 1.34 275.23 ± 1.05 346.21 ± 1.04
Polygalacturonase (PG) 68.90 ± 1.36 270.75 ± 0.06 365.06 ± 0.50
Pectinmethylesterase (PME) 60.86 ± 0.21 275.77 ± 0.00 354.72 ± 0.53
Laccase 59.81 ± 1.26 275.98 ± 1.24 347.55 ± 1.24

Flax Control 65.91 ± 0.78 266.98 ± 0.00 364.35 ± 0.97

Xylanase + cellulase 68.75 ± 0.71 272.08 ± 0.48 366.05 ± 0.48
Xylanase 70.58 ± 2.39 268.57 ± 1.05 368.60 ± 0.28
Polygalacturonase 67.04 ± 4.29 267.33 ± 6.58 361.56 ± 0.91
Pectinmethylesterase 64.59 ± 5.47 267.32 ± 3.37 361.62 ± 2.88
Laccase 66.54 ± 0.62 264.74 ± 0.40 362.01 ± 0.40
When compared to the control, all systems were significantly different at (p < 0.05), except flax treated with PG and PME.
For enzyme treated hemp, only xylanase + cellulase and PG were significantly different at (p < 0.05).
Laccase, xylanase, xylanase + cellulase treated hemp and laccase, xylanase treated flax were significantly different from the control at (p < 0.05).

Flax fibres treated with xylanase + cellulase were characterized 4. Conclusion

by lower thermal resistance when compared to the control exper-
iments. Flax fibres typically have more hemicellulose than hemp Enzymes can be effectively used to remove the hygroscopic
fibres and therefore the removal of xylan resulted in a weaker pectic and hemicellulose material producing a more homogenous
structure (Table 2). Further evidence in support to this explana- fibre surfaces with improved thermal properties. From a struc-
tion was the increased maximum degradation temperature which tural standpoint, the removal of these components resulted in an
could be attributed to the removal of the least stable hemicellu- increase in individual bundle exposure, as observed from the SEM
losic fraction. Flax fibres treated with laccase were characterized micrographs. The ultrastruture of flax fibres allows for more effi-
by a lower maximum degradation and onset of degradation tem- cient enzymatic treatment compared to the hemp fibres making
peratures. The removal of lignin located in the interstitial spaces of it a superior feedstock option for reinforcing polymeric matrices.
cellulose and hemicellulose is known to act as an adhesive hold- Enzymes can be used to address the limited thermal stability of
ing the fibrils network together the network together (Vila et al., natural fibres by selectively removing the least stable pectic and
2011). Laccase removes lignin, which correlated to a decrease in hemicellulosic content. In summary, enzymatic treatment of nat-
these properties because the structure is less rigid. In summary, ural fibres such as hemp and flax can be used to improve both
enzymes can be used to selectively degrade components that limit surface and thermal properties for subsequent modifications and
thermal stability. The pectinases had no significant (p < 0.05) effect incorporation in polymeric matrices.
on the thermal properties of the flax fibres and will not be discussed
further. Acknowledgements
To our best knowledge, this is the first report detailing the use of
enzymatic treatments to improve the thermal properties of natural The authors would like to acknowledge the financial support of
fibres for composite applications. the Alberta Livestock and Meat Agency. The authors are grateful
M. George et al. / Industrial Crops and Products 53 (2014) 365–373 373

to Alberta Biomaterials Development Centre and Novozymes for Karaduman, Y., Gokcan, D., Onal, L., 2013. Effect of enzymatic pretreatment on the
kindly supplying the materials needed for this study. The experi- mechanical properties of jute-fiber reinforced polyester composites. J. Comp.
Mater. 47 (10), 1–10.
mental support provided by Drs. Shihong Xu and Dimitre Karpuzov Kardas, I., Symonowicz, B., Sztajnowski, S., 2009. Comparison of the effect of PET
from the Alberta Centre for Surface Engineering and Science fibers surface modification using enzymes and chemical substances with respect
(ACSES) at the University of Alberta is much appreciated. The help to changes in mechanical properties. Fibers Textiles Eastern Europe 17 (4),
with SEM imaging provided by Gayle Hatchard at the Department Kharazipour, A., Huettermann, A., Luedemann, H.D., 1997. Enzymatic activation of
of Chemical and Materials Engineering at the University of Alberta wood fibres as a means for the production of wood composites. J. Adhes. Sci.
is also much appreciated. Technol. 11 (3), 419–427.
Li, Y., Pickering, K.L., 2008. Hemp fibre reinforced composites using chelator and
enzyme treatments. Comp. Sci. Technol. 68 (15–16), 3293–3298.
References Liu, L., Cheng, L., Huang, L., Yu, J., 2012. Enzymatic treatment of mechanochemical
modified natural bamboo fibers. Fibers Polym. 13 (5), 600–605.
Anyakora, A., Abubakre, O., 2011. Effect of silane treatment on the impact strength Milanese, A.C., Cioffi, M.O.H., Voorwald, H.J.C., 2012. Thermal and mechanical
properties of oil pal empty fruit bunch fiber reinforced polyester composites. behaviour of sisal/phenolic composites. Composites Part B 43 (7), 2843–2850.
World J. Eng. Pure Appl. Sci. 1 (2), 40–46. Mohanty, A., Misra, M., Drzal, L., 2002. Sustainable bio-composites from renewable
Aracri, E., Vidal, T., 2011. Xylanase- and laccase-aided hexenuronic acids and lignin resources: opportunities and challenges in the green materials world. J. Polym.
removal from specialty sisal fibres. Carbohydr. Polym. 83 (3), 1355–1362. Environ. 10, 19–26.
Aranberri, A., Lampke, T., Bismarck, A., 2003. Wetting behavior of flax fibres as Mukhopadhay, S., Fangueiro, R., 2009. Physical modification of natural fibers and
reinforcement for polypropylene. J. Colloid Interface Sci., 580–589. thermoplastic films for composites—a review. J. Thermoplast. Compos. Mater.
Baley, C., Busnel, F., Grohens, Y., Sire, O., 2006. Influence of chemical treatments on 22 (2), 135–161.
surface properties and adhesion of flax fibre-polyester resin. Composites Part A Mwaikambo, L., Ansell, M., 2002. Chemical modification of hemp, sisal, jute, and
37 (10), 1626–1637. kapok fibers by alkalization. J. Appl. Polym. Sci. 84 (12), 2222–2234.
Baracat, M., Valentim, C., Muchovej, J., Silva, D., 1989. Selection of pectinolytic fibres Palanivelu, P., 2006. Polygalacturonases: active site analyses and mechanism of
for degumming. Biotechnol. Lett. 11, 809–902. action. Indian J. Biotechnol. 5, 148–162.
Bateman, D.F., Basham, H.G., 1976. Degradation of plant cell walls and membranes Pickering, K., Li, Y., Farrell, R., Lay, M., 2007. Interfacial modification of hemp fiber
by microbial enzymes. Encyclopedia Plant Physiol. 4, 316–355. reinforced composites using fungal and alkali treatment. J. Biobased Mater. Bio.
Biagiotti, J., Puglia, D., Kenny, J., 2004. A review on natural fibre-based composites- 1, 109–117.
Part 1. J. Nat. Fibers 1 (2), 37–68. Pietak, A., Korte, S., Tan, E., Downard, A., Staiger, M.P., 2007. Atomic force microscopy
Bismarck, A., Mohanty, A., Aranberri, I., Ibon, A., Syliva, C., Manjusri, M., Georg, H., characterization of the surface wettability of natural fibres. Appl. Surf. Sci. 253
Jürgen, S., 2001. Surface characterization of natural fibers; surface properties (7), 3627–3635.
and the water up-take behavior of modified sisal and coir. Green Chem. 003, Prade, R.A., 1996. Xylanases: from biology to biotechnology. Biotechnol. Genet. Eng.
100–107. Rev. 13, 101–131.
Bledzki, A., Mamum, A., Gabor, M., Gutowski, V., 2008. The effects of acetylation Ramadevi, P., Sampathkumar, D., Srinivasa, C., Bennehalli, B., 2012. Effect of alkali
on properties of flax fibre and its polypropylene composites. Polym. Lett. 2, treatment on water absorption of single cellulosic abaca fibre. BioRes. 7 (3),
413–422. 3515–3524.
Boisset, C., Chanzy, H., Schulein, M., Henrissat, 1997. An ultra structural study of the Reddy, N., Yang, Y., 2005. Biofibres from agricultural byproducts for industrial appli-
interaction of a fungal endoglucanase from Humicola insolen with cotton fibres. cations. Trends Biotechnol. 23 (1), 22–27.
Cellulose 4, 7–20. Saleem, Z., Rennebaum, H., Pudel, F., Grimm, E., 2007. Treating bast fibers with
Boisset, C., Petrequin, C., Chanzy, H., Henrissat, B., Schulen, M., 2001. Optimized pectinase improves mechanical characteristics of reinforced thermoplastic com-
mixtures of recombinant humicola insolens cellulases for the biodegradation of posites. Comp. Sci. Technol. 68 (2), 471–490.
crystalline cellulose. Biotechnol. Bioeng. 72 (3), 339–345. Sgriccia, N., Hawley, M., Misra, M., 2008. Characterization of natural fiber surfaces
Brigida, A., Calado, V., Goncalves, L., Coelho, M., 2010. Effect of chemical treatments and natural fiber composites. Composites Part A. 39 (10), 1632–1637.
on properties of green coconut fibre. Carbohydr. Polym. 79, 832–838. Tajvidi, M., Takemura, A., 2008. Effect of fibre content and type, compatabilizers
Fei, Y., Wu, Q., Zhou, D., 2009. Thermal decomposition of natural fibres: Global kinetic and heating rate on thermogravimetric properties of natural fibre high density
modeling with nonisothermal thermogravimetric analysis. J. Appl. Polym. Sci. polyethylene composites. Polym. Comp. 30 (9), 1226–1233.
114 (2), 834–842. Tserki, V., Zafeiropoulos, N., Simon, F., Panayiotou, C., 2005. A study of the effect of
Gassan, J., Gutowski, V., 2000. Effects of corona discharge and UV treatment on acetylation and propionylation on surface treatments of natural fibres. Compos-
the properties of jute-fibre epoxy composites. Compos. Sci Technol. 60 (15), ites Part A 36 (8), 1110–1118.
2857–2863. Vila, C., Barneto, A.G., Fillat, A., Vidal, T., Ariza, J., 2011. Use of Thermogravimet-
Gonzalez, A.U.J., Olayo, R., Franco, P., 1999. Chemical modification of henequén fibers ric analysis to monitor the effects of natural laccase mediators on flax pulp.
with an organosilane coupling agent. Composites Part B 30 (3), 321–331. Bioresour. Technol. 83 (3), 1355–1362.
Gulati, D., Sain, M., 2006. Fungal modification of natural fibers: a novel method Wiener, J., Kovacic, V., Dejlova, P., 2003. Differences between flax and hemp. Autex
of treating natural fibers for composite reinforcement. J. Polym. Environ. 14, Res. J. 3 (2), 58–63.
347–352. Williams, P., Sobering, D., Antoniszyn, J., 1998. Protein testing methods at the Cana-
Gustavsson, M.T., Persson, P.V., Iversen, T., Martinelle, M., Hult, K., Teeri, T.T., Brumer, dian Grain Commission. In: Proceedings of the Wheat Protein Symposium,
H., 2005. Modification of cellulose fibre surface by use of a lipase and a xyloglucan Saskatoon, SK Canada.
endotransglycoslyase. Biomacromolecules 6 (1), 196–203. Williams, T., Hosur, M., Theodore, M., Netravali, A., Rangari, V., Jeelani, S., 2011.
Gutierrez, A., Rencoret, J., Cadena, E., Rico, A., Barath, D., Rio, J., 2012. Demonstration Time effect on morphology and bonding ability in mercerized natural fibres for
of laccase-based removal of lignin from wood and non-wood plant feedstock. composite reinforcements. Int. J. Polym. Sci. 2011, 1–9.
Bioresour. Technol. 119, 114–122. Yang, H., Yan, R., Chen, H., Lee, D., Zheng, C., 2007. Characteristics of hemicellulose,
Jayabal, S., Sathiyamurthy, S., Loganathan, K., Kalyanasundaram, S., 2012. Effect of cellulose and lignin pyrolysis. Fuel 86 (12–13), 1781–1788.
soaking time and concentration of NaOH solution on mechanical properties of Yuan, X., Jayaraman, K., Battacharyya, D., 2004. Effects of plasma treatment in
coir-polyester composites. Bull. Mater. Sci. 35, 567–574. enhancing the performance of wood fibre-polypropylene composites. Compos-
Kalia, S., Kaith, B., Kaur, I., 2009. Pretreatments of natural fibres and their applica- ites Part A 35 (12), 1363–1374.
tion as reinforcement in polymer composites, a review. Polym. Eng. Sci. 49 (7), Zafeiropoulos, N., Vickers, P., Baillie, C., 2003. An experimental investigation of mod-
1253–1272. ified and unmodified flax fibres with XPS, ToF-SIMS and ATR-FTIR. J. Mater. Sci.
Katapodis, P., Vrsanska, M., Kekos, D., Nerinckx, W., Biely, P., Claeyssens, M., 2003. 38, 3903–3914.
Biochemical and catalytic properties of an endoxylanase purified from the cul-
ture filtrate of Sporotichum thermophile. Carbohydr. Resis. 338 (18), 1881–1890.