Isolation and characterization of cellulose nanofibrils from wheat straw

Carbohydrate Research 346 (2011) 76–85

Contents lists available at ScienceDirect

Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres

Isolation and characterization of cellulose nanofibrils from wheat straw
using steam explosion coupled with high shear homogenization
Anupama Kaushik a,⇑, Mandeep Singh b
a
b

University Institute of Chemical Engineering and Technology, Panjab University, Chandigarh, India
Centre for Emerging Areas in Science & Technology, Panjab University, Chandigarh, India

a r t i c l e

i n f o

Article history:
Received 17 July 2010
Received in revised form 20 October 2010
Accepted 26 October 2010
Available online 30 October 2010
Keywords:
Wheat straw
Cellulose nanofibers
High shear homogenizer
Steam explosion
AFM

Thermal degradation

a b s t r a c t
Cellulose nanofibrils of diameter 10–50 nm were obtained from wheat straw using alkali steam explosion
coupled with high shear homogenization. High shear results in shearing of the fiber agglomerates resulting in uniformly dispersed nanofibrils. The chemical composition of fibers at different stages were analyzed according to the ASTM standards and showed increase in a-cellulose content and decrease in
lignin and hemicellulose. Structural analysis of steam exploded fibers was carried out by Fourier Transform Infrared (FT-IR) spectroscopy and X-ray diffraction (XRD). Thermal stability was higher for cellulose
nanofibrils as compared to wheat straw and chemically treated fibers. The fiber diameter distribution
was obtained using image analysis software. Characterization of the fibers by AFM, TEM, and SEM showed
that fiber diameter decreases with treatment and final nanofibril size was 10–15 nm. FT-IR, XRD, and TGA
studies confirmed the removal of hemicellulose and lignin during the chemical treatment process.
Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction
Natural fibers are abundantly present in plants such as grasses,
reeds, stalks, and woody vegetation. They are also referred to as
cellulosic fibers due to the main chemical component cellulose,
or as lignocellulosic fibers, since the fibers usually contain a natural
polyphenolic polymer, lignin, in their structure. These fibers are
long regarded as being promising candidates for replacing conventional reinforcing fibers (e.g. glass fibers) in composites for
semi-structural and even structural applications. The biodegradable nature of plant fibres can contribute to the formation of a
healthy eco-system as well as their high performance fulfills the
economic interest of industries. Plant based natural fibres like
sisal,1 jute,2 bamboo,3 wood4,5, and paper in their natural condition, as well as several waste cellulose products, such as shell flour,
wood flour, and pulp, can be used as reinforcement materials for
different thermosetting and thermoplastic polymeric matrices.
Cellulose nanofibers have gained importance due to their unique characteristics such as very large surface to volume ratio, high
surface area, good mechanical properties including a high Young’s
modulus, high tensile strength6,7 and a very low coefficient of thermal expansion,8 and formation of highly porous mesh as compared
to other commercial fibers. Functional hydroxyl groups in cellulose
also enable chemical modifications for further applications.

⇑ Corresponding author. Tel.: +91 0172 2534925; Mob.: +91 9815177772.
E-mail addresses: , (A.
Kaushik), (M. Singh).
0008-6215/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2010.10.020

Biocompatibility, non-toxicity, and biodegradability of cellulose
nanomaterials are important properties in biochemical and biomedical applications. All these features make cellulose microfibrils
a very promising material for nanotechnology.
Plant cell walls usually consist of rigid cellulosic microfibrils
embedded in soft hemicelluloses and lignin matrix. It is the structural material of the fibrous cells with high levels of strength and
stiffness per unit weight. Cellulose has a straight carbohydrate
polymer chain consisting of b-(1?4)-linked glucopyranose units
and a degree of polymerization (DP) of about 10,000.9 Hydroxyl
(–OH) groups in cellulose structures play a major role in governing
its reactivity and physical properties. Several groups have reported
methods for preparing cellulosic microfibrils.10–16 These materials,
however, are rather non-homogeneous and in addition to microfibrils contain larger fibril bundles and residual fiber fragments.
Steam explosion is one of the ways of fibril isolation, which is to
be subsequently used in chemical fractionation and biotechnological conversion. The steam explosion process was first introduced
by Mason in 1927 to defibrate wood into fiber for broad production. High pressure steaming followed by rapid decompression is
called steam explosion. The steam explosion process includes saturating the dry material with steam at elevated pressure and temperature followed by sudden release of pressure, during which the
flash evaporation of water exerts a thermo mechanical force causing the material to rupture.17 Steam explosion in alkaline medium
results in the hydrolysis of hemicellulose within the fiber, and the
resulting sugars can be subsequently washed out in water, leaving
a residue of R-cellulose and lignin.18 It also leads to a cleavage of


A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

hemicellulose–lignin bonds. The reaction results in an increased
water solubilization of hemicelluloses and in an increased solubility of lignin in alkaline or organic solvents, leaving the cellulose as
a solid residue with a reduced degree of polymerization.17 The
advantages of steam explosion include a significantly lower environmental impact, low energy consumption, lower capital investment, and less hazardous process chemicals.
In this study, an effort was made to isolate cellulose nanofibrils
from wheat straw using steam treatment with subsequent explosive defibrillation. Steam explosion in alkaline medium followed
by hydrochloric acid treatment and high shear homogenization
was found to be effective in the depolymerization and defibrillation of the fiber to produce cellulose nanofibrils. With this method,
the substrate, that is, wheat straw, is loaded into a pressure vessel
and heated by steam injection for a defined time-temperature period. At the end of this period of heat treatment, the pressure drop is
suddenly made. A rapidly opening valve was used so that, after
treatment, the contents of the reaction vessel were almost instantaneously depressurized and discharged through a nozzle. Alkaline
medium has been used for the explosive treatment. This is followed by bleaching and acidic treatment as well as high shear
homogenization leading to breaking of agglomerates. These cellulose nanofibers obtained were characterized by AFM, TEM, SEM,
FT-IR, XRD, and TGA.

2. Experimental
2.1. Material
Wheat straws were obtained from neighboring fields. They
were thoroughly washed to remove any extraneous impurities
and dried before use. This is extremely low cost material, which
is used for cattle feed. Other chemicals used in the experiments, sodium hydroxide (NaOH), hydrochloric acid (HCl) and hydrogen
peroxide (H2O2) were supplied by Merck India Pvt. Ltd.
2.2. Extraction of cellulose nanofibrils from wheat straw
2.2.1. Chemical treatment
2.2.1.1. Preparation of steam exploded fibers.
The preparation was done in two steps. First, wheat straw fibers of length
around 2–5 cm were soaked in 2% solution of NaOH for overnight
and then treated in 10–12% NaOH solution (solid to liquid ratio
around 1:8) in an autoclave at pressure around 20 bars for 4 h at

200 ± 5 °C. The first treatment removed excessive impurities from
the surface of the fibers and resulted in swelling of fibers thereby
making further treatments easy. Second, treatment removed
excessive amount of lignin from the fibers. The obtained wheat
straw pulp was then washed several times in distilled water till
it was free of alkali.
2.2.1.2. Preparation of steam exploded bleached fibers.
Alkaline treated pulp was then soaked in 8% solution of H2O2 (v/v) and
kept overnight to remove any residual lignin and hemi cellulose that
may have been present.
2.2.1.3. Treatment of steam exploded bleached banana fibers in
acidic medium.
Bleached pulp was treated with 10% HCl
(1 N) solution and kept in Branson 2510 E-DTH ultrasonicator at
temperature around 60 ± 1 °C for 5 h. Finally the fibers were taken
out and washed several times with distilled water in order to neutralize the final pH and then dried. Hydrochloric acid has been used
for hydrolysis instead of sulfuric acid, which is a common choice as
it assists in dispersion and separation of nanofibrils due to introduction of sulfate ester groups randomly on the surface resulting

77

in nonflocculating suspensions. For composite applications, the
sulfate groups are problematic due to decreased thermal stability
after drying, precluding typical polymer melt processing. Battista
and Smith19 discovered that formation of stable suspensions can
also be achieved by hydrolysis of cellulose using hydrochloric acid
followed by mechanical disintegration.
2.2.2. High shear treatment of chemically treated fibers
Finally, the fibers were suspended in water and continuously
stirred with a Fluko FA25 high shear homogenizer for 15 min.

The high shearing action breaks down the fiber agglomerates and
results in nanofibrils.
2.3. Characterization of cellulose nanofibrils
2.3.1. Morphology of cellulose nanofibrils
AFM imaging was used to characterize the dimensions and
homogeneity of cellulose nanofibrils obtained after chemical and
mechanical treatment. AFM was done in tapping mode wherein
the oscillation frequency was constant and the changes in amplitude monitored. During each oscillation cycle the tip is briefly in
contact with the surface and the interaction forces between the
sample and the tip cause a reduction in amplitude.
The images were scanned in tapping mode in air using silicon
cantilevers (Bioscope II AFM, VEECO) and the drive frequency of
the cantilever was about 200–300 kHz with scan rate of 0.5–3 Hz
(usually around 2 Hz). The sample size taken is 10 lm  10 lm
(Fig. 2a), 1.5 lm  1.5 lm (Fig. 2b), 5 lm  5 lm (Fig. 2c), and
3 lm  3 lm (Fig. 2d) while the z scale is 100 nm. No image processing except flattening was made. Samples were fixed on metal
discs with double-sided adhesive tape. To eliminate external vibration noise, the microscope was placed on an active vibrationdamping table.
The diameter and length of cellulose nanofibers (CNFs) extracted from wheat straw were examined by using transmission
electron microscope (TEM) model Hitachi-2100. Images were taken at 80 kV accelerating voltage. A drop of a dilute aqueous cellulose nanofiber suspension was deposited on the carbon-coated
grids and allowed to dry at room temperature.
Scanning electron microscope model JSM JEOL-6490 was used
for microstructural analysis of cellulose microfibers obtained after
steam explosion. Samples were mounted on a metal stub and platinum coated by using sputter coating technique for 20 s to make
them conducting. Images of fibers were taken at 20 kV accelerating
voltage at different magnifications.
2.3.2. Chemical characterization of the nanofibers
Chemical composition of fibers was estimated according to the
following ASTM procedures: a-cellulose (ASTM D1103-55T), lignin
(ASTM D1106-56) and holocellulose (ASTM D1104-56). The standard deviations were calculated by conducting several replicate
measurements for each sample.

The a-cellulose, hemicelluloses and lignin content were calculated by Eqs. 1–3 as follows:
weight of oven dry a À cellulose residue
 100
ð1Þ
W ÂP



weight of oven dry holocellulose residue
 100 À A
Hemicellulose percentage ¼
W ÂP

a-cellulose percentage ðAÞ ¼

ð2Þ
weight of oven dry lignin residue
Lignin percentage ¼
 100
W ÂP

ð3Þ

where W is the weight of the original oven dry fiber sample, and P is
the proportion of moisture-free content.


78

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

2.3.3. Fourier Transform Infrared (FTIR) spectroscopy
FT-IR analysis of raw as well as chemically treated wheat straw
fibers was done in order to obtain composition of the fibers before
and after treatment. A Perkin Elmer RX infrared spectrophotometer
was used to obtain spectra. Fibers were ground and mixed with
KBr (sample/KBr ratio, 1/99) to prepare pastilles. FT-IR spectra
were recorded in a spectral range of 4000–450 cmÀ1 with a resolution of 2 cmÀ1, taking four scans for each sample.
2.3.4. X-ray diffraction
X-ray diffraction (WAXRD) profiles of wheat straw fibers before
and after chemical process were collected in order to examine the
crystallinity of the samples. The samples were taken in powdered
form and analyzed by using a Philips X’Pert Pro X-ray diffractometer system. The radiation was Cu Ka (k = 1.54060 Å) with 40 kV
voltage and 40 mA intensity.
Crystallinity of cellulose in pulp samples was calculated from
diffraction intensity data. The major diffraction planes of cellulose
 0 2 1, 0 0 2, and 0 4 0 are present at 14.8°, 16.7°,
namely 1 0 1, 1 0 1,
20.7°, 22.5°, and 34.6° 2h angle, respectively.20 The crystallinity index was obtained using the Eq. 421:

Crystallinity index ¼ 100 Â

I0 0 2 À IAmorph
I0 0 2

ð4Þ

where I0 0 2 is the maximum intensity of the (0 0 2) lattice diffraction and IAmorph is the intensity diffraction at 18° 2h degrees. The
calculation of the X-ray crystallinity order index was also performed using Eq. 5. The crystallinity index was calculated from
the fraction of the ratio of the (0 0 2) to the sum of (1 0 1), (0 2 1)

and (0 0 2) refraction areas:

Crystallinity order index ¼

A0 0 2
A1 0 1 þ A1 0 1 þ A0 0 2

ð5Þ

The Eqs. 5 and 6 were used for the amorphous fraction in terms
of intensity and area to calculate the crystallinity index. The selected position that was assigned to the amorphous fraction was
the diffraction angle at around 18° 2h degree.
The crystallite size or the thickness of crystal in a direction perpendicular to its Miller plane was estimated using the Scherrer
equation. This is a method based on the width of the diffraction
pattern in the X-ray reflected crystalline region.

t hkl ¼

Kk
bhkl cos h

ð6Þ

In Eq. 6, thkl is the thickness of crystallites at the (hkl) plane of
diffraction, k is an X-ray wavelength (k = 0.1542 nm for Cu Ka), h
is the Bragg angle of the reflection, bhkl is the pure integral of width
of the reflection at half maximum height, and K is the Scherrer constant that falls in the range 0.87–1.0.22 The crystallite size of cellulose nanofibrils was determined by using the diffraction pattern
obtained from 0 0 2 lattice plane.
2.3.5. Thermal characterization
Thermogravimetric analysis was undertaken to compare the

degradation characteristics of the chemically treated fibers with
the untreated ones. The thermal stability of each sample was
determined using a thermogravimetric analyzer (TGA) of type PERKIN ELMER STA-6000. It has weighing capacity up to 1500 mg with
resolution around 0.1 lg. Samples were taken in a very small quantity (in milligram) in a sample cup made of alumina and having
maximum capacity around 180 lL. All the tests were performed
in nitrogen environment and at heating rate of 10 °C/min from
room temperature to 600 °C.

2.3.6. Fiber diameter measurements
Fiber diameter measurements of the wheat straw fibers after
chemical and mechanical treatment were undertaken using a
UTHSCSA Image Tool image analyzer program (IT version 3). TEM
images of the cellulose nanofibers were used to measure the diameters. The images were loaded into the software and diameters of
the fibers were measured using a two point measuring analysis.
The scale of the software was calibrated using the scale bars on
each TEM image. Approximately, 300 measurements were taken
to obtain each fiber diameter distribution.
2.3.7. Degree of polymerization
Viscosity of the cellulosic preparations after steam explosion,
bleaching, and acid treatment was determined by British Standard
Method for determination of limiting viscosity number of cellulose
in dilute solution: Part 1: Cupric–ethylenediamine method (BS
6306: Part 1: 1982). The viscosity average Degree of Polymerization (DOP) of the cellulose samples was estimated by
DOP0.90 = 1.65 [g]. Molecular weight of the cellulosic preparations
was then calculated from their DOP by multiplying with 162, the
molecular weight of an anhydroglucose.23

3. Results and discussion
3.1. Morphology and chemical characterization
The alkaline steam explosion results in structural as well as

chemical changes on fiber surfaces. SEM pictures of the wheat
straw after steam explosion were taken to investigate the structure
of these fibers. The SEM micrographs are shown in Figure 1a–d.
These clearly show individual fibers after the removal of hemicelluloses, lignin, and pectin after chemical treatment, which are the
cementing materials around the fiber-bundles. It is clear from the
pictures that the average diameters of the fibers are about 10–
15 lm, which is lower than the average size of fiber bundles before
chemical treatment. Reduction in particle size because of the dissolution of the hemicelluloses and lignin is clearly supported by
chemical analysis and FT-IR data as given in Tables 1 and 2.
Figure 2 shows the phase images of AFM of cellulose nanofibrils
at different magnifications. Figure 1a and b are phase images and
Figure 1c and d are topographical images of the nanofibers extracted. The results reveal that fiber diameter has reduced to the
nanometer range after chemical and mechanical treatment. From
the AFM photographs, it is clear that the fibers are found to be
slightly agglomerated. The surface of the fibers is found to be
smooth from the AFM images in contrast to rough surface of microfibrils as seen in SEM images.
Figure 3(i) shows TEM images of the cellulose nanofibers after
the chemical and mechanical treatments. Mechanical treatment
resulted in defibrillation of the cellulose nanofibers from the cell
wall and TEM images reveal the separation of these nanofibers
from the microsizes fiber bundles. The diameter of the fibers is
found in the range of 10–60 nm. A tendency of agglomeration
could also be observed from TEM. It is not clear whether this was
due to drying of the suspension onto the carbon film covering
the carbon grids or if it reflected the state of the suspension. The
average diameter was calculated from the electron micrographs
using digital image analysis software, UTHSCSA. Most of the particles were found in the diameter range of 30–40 nm.
Figure 3(ii) shows the distribution of nano-fiber diameter after
final treatment. Only 3% of fibers have diameter >70 nm. Almost
64% of fibers have diameter between 30–50 nm, 19% of the fibers

have a diameter less than 30 nm.
Table 1 shows the chemical composition of raw, alkaline steam
exploded, bleached fibers and acidic treated fibers. Raw fiber has


A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

79

Figure 1. SEM image of the steam-exploded micro-structured cellulose fibers at different magnifications.

Table 1
Chemical composition of untreated and steam exploded and chemically treated
wheat straw
Material

Percentage of acellulose

Percentage of hemi
cellulose

Percentage of
lignin

Untreated wheat
straw
Alkaline steam
exploded
Acid treated
fibers

Cellulose nanofibers

45.70 ± 0.18

37.12 ± 0.9

17.43 ± 2.1

65.29 ± 2.51

22.22 ± 1.12

10.27 ± 1.67

75.28 ± 2.37

12.34 ± 1.18

8.12 ± 1.35

86.38 ± 3.12

8.13 ± 0.8

6.34 ± 1.25

Table 2
Bragg angle, fiber diameter, crystallinity index and crystallinity order index
Fiber
type

Bragg
angle (°)

Fiber diameter
(nm)
(Eq. 6)

Crystallinity
index (%)
(Eq. 4)

Crystallinity
order index
(Eq. 5)

WS
CTWS
CNF

11.259
11.23
11.35

—
5.23
3.91

54.42
66.60

79.87

57.43
71.33
80.05

in alkaline medium, the hemicelluloses and lignin components
present in the raw fiber will dissolve out.
Yamashiki et al.24 proposed an explanation for the solubility of
steam exploded cellulose in NaOH solution, suggesting that during
the steam explosion there is a partial breakdown of the intramolecular hydrogen bond at the C-3 and C-6 positions of the glucopyranose unit and this results in significant variations in the network
and strength of the hydrogen bonds of the cellulose hydroxyls.
However, the complete removal of these components does not take
place. During the explosion, some changes occur in the arrangement of macromolecular chains. Xiao et al.25 proposed that during
steam explosion, the hemicellulose is partially hydrolyzed and the
lignin is depolymerized, giving rise to sugars and phenolic compounds that are soluble in water. The hydrolyses of glycosidic linkages in hemicellulose and the ether linkages in lignin are catalyzed
by acetic acid formed at high temperature from acetyl groups present in hemicellulose (autohydrolysis). At the end of the process, the
steam was suddenly released providing additional mechanical defibrillation. The cellulose is de-polymerized and defibrillated resulting in crystalline nanofibrils. The removal of hemicelluloses and
lignin in chemically treated fibers is confirmed in FT-IR results also.
3.2. Fourier Transform infrared (FT-IR) spectroscopy

the highest percentage of hemicellulose and lignin and the lowest
percentage of a-cellulose as compared to treated fibers. The acellulose content increases from 45.7% in raw fiber to 86.38% in
finally treated fiber. Similarly, hemicelluloses content decreases
from 37.12% to 8.13% and lignin decreases from 17.43% to 6.34%.
When the steam explosion process is done, there is a decrease in
the hemicellulose and lignin component present in the wheat fiber.
This shows that during steam explosion, substantial breakdown of
the lignocellulosic structure, partial hydrolysis of the hemicellulosic fraction, and depolymerization of the lignin components have
occurred.17 When the raw fiber is subjected to steam explosion

Untreated and chemically treated wheat straw fibers were analyzed using FT-IR to examine the changes occurring in the chemical
constituents of fibers before and after the chemical treatments.
Figure 4 shows the FT-IR spectrum of the raw wheat straw fibers
and chemically treated wheat straw fibers. The peaks in area
3369 cmÀ1 correspond to O–H stretching band, that is, due to
vibrations of the hydrogen bonded hydroxyl group25–27 The peaks
at 2922 cmÀ1 are due to the aliphatic saturated C–H stretching
vibration in lignin polysaccharides (cellulose and hemicelluloses).
The hydrophilic tendency of raw fibers and chemically treated
wheat straw fibers is reflected in the broad absorption band in


80

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

Figure 2. AFM images of cellulose nanofibrils (tapping mode): (a) 10 lm  10 lm, (b) 1.5 lm  1.5 lm phase images, (c) 5 lm  5 lm topographical image and (d)
3 lm  3 lm topographical image.

the 3700–3100 cmÀ1 region, which is related to the –OH groups
present in their main components. Peak at 1734 cmÀ1 in the untreated wheat straw is attributed to either the acetyl and uronic ester groups of the hemicelluloses or the ester linkage of carboxylic
group of the ferulic and p-coumeric acids of lignin and/or hemicelluloses.17,26,28 It can be seen in Figure 4 that this peak is almost absent in the spectra of the chemically treated fibers, which indicates
the near cleavage of these ester bonds. The peak at 1652 cmÀ1 may
be due to the bending mode of the absorbed water and some contributions from carboxylate groups.26 The aromatic C@C stretch
from aromatic ring of lignin gives two peaks at 1510 and
1426 cmÀ1 that can be observed in untreated wheat straw
fibers.17,28,29
The peak at 1510 cmÀ1 has almost vanished and the intensity of
peak at 1426 cmÀ1 has significantly decreased in chemically treated fibers attributing to partial removal of lignin. The peaks at

1373 cmÀ1 represent C–H asymmetric deformation. The intensity
of the peak at 1258 cmÀ1 has sharply decreased after chemical
treatment indicating the removal of hemicelluloses. The region of
1200–1059 cmÀ1 represents the C–O stretch band and deformation
bands in cellulose, lignin and residual hemicelluloses.26 The
increase of band at 897 cmÀ1 in chemically treated wheat straw fibers indicates the typical structure of cellulose (due to b-glycosidic
linkages of glucose ring of cellulose).30
3.3. X-ray diffraction
WAXRD analyses of untreated and treated fibers were done in
order to study the crystalline behavior of the fibers and to assess

the relationship between structure and properties of fiber. Cellulose show crystalline nature while lignin is amorphous in nature.
As a result, the crystallinity of the fibers should improve after removal of the lignin. Based on this idea X-ray diffraction of powdered samples of the untreated wheat straw fibers, chemically
treated fibers, and mechanically high shear treated fibers was carried out so as to examine the changes occurring in the crystalline
nature of the wheat straw fibers after the chemical treatment.
Figure 5(i) shows the XRD profiles of the untreated wheat straw fibers, chemically treated fibers, and cellulose nanofibrils after
mechanical shear treatment.
As discussed earlier by many authors, lignocellulosic fibers are
composed of three major components namely cellulose, hemicellulose, and lignin. Crystalline microfibrils of cellulose are surrounded
by amorphous hemicellulose and the whole is embedded in the
matrix of lignin. Crystalline structure of cellulose and hemicellulose exhibits variability in both structure and constitution. In native cellulose the length of the crystallites can be 100–250 nm
with average cross-sections of 3–10 nm. Chemical and mechanical
treatments affect the crystallite size as well as the crystallinity of
cellulose.
From Figure 5(i) it is clear that finally treated cellulose nanofibrils show crystalline nature. The peak intensity at 2h = 22.6 corresponding to 0 0 2 lattice plane increases with the chemical
treatment. It further increases sharply with high shear mechanical
treatment. Chemically treated fibers show a narrow peak at 26.5°,
which may be due to heavy loading of chemicals.
Higher crystallinity is due to more efficient removal of noncellulosic polysaccharides and dissolution of amorphous zones. The

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

81

Figure 3(i). TEM images of cellulose nanofibrils at different magnifications: (a) 50,000Â, (b) 15,000Â, (c) 80,000Â and (d) 40,000Â.

results also confirm that hydrolysis takes place preferentially in the
amorphous region as acids dissolve the amorphous regions while
crystalline regions are more stable toward chemical attack. This increase of crystallinity after acid treatment has been reported by
several authors.17,28,31
To evaluate the crystallinity order index, peak separation of
each X-ray diffraction was conducted for each plane. The operation
used a non-linear multi-peak fitting function of OriginPro 8 software. The Pseudo-Voigt type I equation was chosen. There are five
variables in this equation; peak offset (y), peak amplitude or area
(A), peak center (xc), peak width (w) and peak shape factor (mu).
After an equation and data from WAXD were assigned, the fitting

was initiated by entering number of replicas equalent to the number of peaks À 1. The fitting was performed with several iterations
until the optimum fit result was obtained. This was indicated by no
further decline of v2 value and R2 approached greater than 0.99.
Fitted data for untreated wheat straw fiber as an example have
been shown in Figure 5(ii).
Table 2 gives the value of Bragg angle, crystallite size (t), crystallinity index (CI), and crystallinity order index (COI) for untreated, chemically treated, and mechanically treated fibers. The
crystallinity increases from 54.42% for untreated fibers to 79.87%
for cellulose nanofibrils and crystallinity order index increases
from 57.43% to 80.05% for cellulose nanofibrils. The increase in


82

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

40
35

25

B

frequency (percent)

30

20
15
10
5
0
0-20

20-30

30-40

40-50

50-60

60-70

>70

particle diameter (nm)
Figure 3(ii). Diameter distribution of wheat straw fiber after chemical treatment
(obtained from TEM analysis).

crystallinity could also be attributed to the facts that steam at high
temperatures reorganizes the amorphous and paracrystalline cellulose regions. It releases strains from native cellulose that arise
during the crystallization phase of cellulose biosynthesis and the
interaction of the cellulose with hemicellulose and lignin in cell
wall formation.17 It is evident from the results that with chemical
and mechanical treatment value of b increases and thus there is a
decrease in fiber diameter.
3.4. Thermal characterization
As we aim to reinforce the particles in thermoplastic starch
based polymers to make completely biocompatible polymers, the
thermal properties of these untreated and treated fibers is important in order to gauge their applicability for biocomposites, in
which processing temperature for thermoplastic polymers rises
above 200 °C.

The TGA results for wheat straw fibers at different stages are
shown in Figure 6(i). These results clearly show that the thermal
stability of the wheat straw fibers increases after chemical treatments and it further increases after high shear mechanical treatments. The degradation temperature increased after chemical
treatment. This was probably because more non-cellulosic material was removed and the high degree of structural order was retained. This revealed a relationship between structure and the
thermal degradation of cellulose. A greater crystalline structure required a higher degradation temperature.32 However, both noncellulosic components and the crystalline order of cellulose played
an important role in thermal degradation of the fibers.33
Different amounts of the residues are obtained from the fibers
remaining after 600 °C heating for untreated, treated and nanofibers. Maximum residue was obtained in untreated wheat straw
fibers and relatively small amount in nanofibers after chemimechanical treatment. It can be concluded that higher temperature

of thermal decomposition and lesser residual mass of the fibers
obtained after chemi-mechanical treatment have been related to
partial removal of hemicelluloses and lignin from the fibers and
higher crystallinity of the cellulose. These results are consistent
with results obtained from crystallinity and FT-IR measurements.
Figure 6(ii) shows DTG for untreated, chemically treated, and
mechanically treated fibers. For untreated wheat straw fibers the
first decomposition shoulder peak at about 236 °C is attributed to
thermal depolymerisation of hemicelluloses or pectin in an inert
atmosphere (mass loss 6.3%)34; (2) the major second decomposition
peak at about 321.86 °C is attributed to cellulose decomposition
(mass loss 27.81%); (3) the small tail peak at 343 °C (mass loss
30%) may be attributed to degradation of lignin35; (4) a very small
peak at about 433 °C is for oxidative degradation of the charred residue. In chemically treated fibers the first peak disappears but the
second peak exists while in nanofibers, DTG shows a sharp peak
only at 337.58 °C indicating decomposition of crystalline cellulose.
Thermal decomposition parameters for untreated, chemically
treated, and mechanically treated fibers were determined from
the TG, DTG, and second time derivatives, D2TG, curves at a heating
rate of 10 °C/min as an example (Fig. 6(iii)) using method given by
Groni et al.36 The extrapolated onset temperature of decomposi-

1

spl.code:CNF
spl.code:Treated
spl.code:WS

0.9
0.8

Transmittance

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
3500

3000

2500

2000

1500

1000

500

Wavenumbers [1/cm]
Figure 4. FT-IR spectra of wheat straw fibers (untreated and chemically treated and cellulose nanofibrils).

0

83

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85
2000

002

1800

Intensity (counts)

1600
1400
1200

101

1000
800

021
101

600

040

400
10

15

20

25

30

35

40

45

50

55

Diffraction angle (2θ)
Figure 5(i). XRD spectra of different fibers: (a) untreated wheat straw, (b)
chemically treated wheat straw, and (c) cellulose nanofibrils.

tion, To, was obtained by extrapolating the slope of the DTG curve
in correspondence with the first local maximum in D2TG curve and
down to the zero level of the DTG axis. The peak temperature, Tp,
was determined by DTG peak where the maximum decomposition
rate was obtained. The percentage weight loss corresponding to
peak temperature is symbolized with MTp. The final, tailing region
indicated the end of cellulose decomposition. Further reactions

continued decomposition of lignin and tar or char was obtained
from main components decomposition.34 The shift temperature,
Ts, is defined here by extrapolating the slope of DTG curve corresponding to the local minimum in D2TG curve in this region and
down to the zero level of DTG axis. The percentage weight loss corresponding to Ts is marked as MTs. The residual solid mass fraction
percentage left at the final temperature, 600 °C is marked as MT600.
The decomposition characteristics of wheat straw, chemically
treated, and mechanically treated fibers are summarized in Table 3.

Figure 5(ii). Wide angle X-ray diffractogram and fitted data of untreated wheat
straw (as an example).

The parameter To indicates an onset decomposition temperature, it
being 239.5 °C for wheat straw, 276.2 °C for chemically treated
straw, and 283.2 °C for cellulose nanofibrils obtained after
mechanical high shear treatment. Weight loss in this period (referred to MTo) was observed around 7% for all fibers. The parameter
Tp presents the maximum decomposition rate of the fibers and it
lies in a range of 321 to 345 °C. The weight loss at this point is indicated by parameter MTp. From peak to shift temperature, fibers
had a rapid degradation in a narrow temperature range. All the fibers completed almost 70% or higher weight loss at 356–372 °C as
shown by MTs and Ts, respectively. The residual weight was maximum for wheat straw which (22%) followed by chemically treated
fibers (18%). Minimum residual weight (11%) was obtained after
mechanical high shear treatment. This is because of absence of
non-organic components in the cellulose nanofibrils as confirmed
by FT-IR and XRD analysis.

Figure 6(i). TG curves: (a) finally mechanical treated fibers, (b) after acidic treatment, and (c) wheat straw.


84

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85

Figure 6(ii). DTG curves: (a) finally mechanical treated fibers, (b) after acidic treatment, and (c) wheat straw.

Figure 6(iii). Determination of decomposition characteristic parameters of wheat straw fibers as an example at a heating rate of 10 °C/min.

3.5. Degree of polymerization
The values for viscosity average degree of polymerization (DOP)
and the corresponding molecular weight of steam-exploded wheat
straw in alkaline medium, bleached pulp, acid treated bleached
pulp, and cellulose nanofibrils were calculated from the intrinsic

viscosity of its solution in 0.5 M CED, according to Robert and
Adrian23 method. For alkaline steam exploded wheat straw, intrinsic viscosity was 720.2 mL/g, viscosity average DOP was 2609.72,
and molecular weight (Mw) was 422,775. For steam exploded
bleached fibers viscosity reduced to 420.3 mL/g, viscosity average
DOP was 1434.54, and molecular weight (Mw) was 232,395. The


85

A. Kaushik, M. Singh / Carbohydrate Research 346 (2011) 76–85
Table 3
Decomposition characteristics of wheat straw, chemically treated fibers and nanofibrils
Fiber
type

To
(°C)

Tp (°C)

Ts
(°C)

Weight loss at To
(MTo) (%)

Weight loss at Tp
(MTp) (%)

Weight loss at Ts
(MTs) (%)

%age residue at T6 0 0
MT6 0 0 (%)

Ts À To
(°C)

MTs À MTo
(%)

WS
CTWS
CNF

239.5
276.2
283.2

321.86
345.71
337.58

356.2
372.3
360.3

6.54
7.35
6.876

27.81
49.88
23.49

63.72
69.94
74.12

78.42
82.45
89.01

116.7
96.1
77.1

57.18
62.59

67.24

immersion of lignocellulosic fibers in dilute alkaline medium facilitates the adhesive nature of the fiber surface by removing natural
and artificial impurities, and causes separation of structural linkages between lignin and carbohydrate and the disruption of lignin
structure. The mechanism of bleaching involves oxidation of lignin,
which leads to lignin dissolution and its degradation. These processes are accompanied by loss of cellulose as evident in decreasing molecular weight and degree of polymerization.
Acid treatment hydrolyzed the traces of hemicelluloses and lignin remaining after the bleaching phase by breaking down the
polysaccharides to simple sugars and hence released cellulose fibers. The combined acid steam treatments effectively reduce the
long micro fibril chains to nanodimensions by maximum explosion
of pressurized steam into the interfibrilar region.17 The viscoisity,
DOP, and Mw further decrease with acidic treatment reaching a
value of viscosity = 123.2 mL/g, DOP = 366.90, and Mw = 59,438. Finally the fibers were subjected to high shear mechanical treatment. The high shearing action breaks down the fiber
agglomerates and results in nanofibrils. The viscoisity of cellulose
nanofibrils obtained after high shear treatment reduced to
92.5 mL/g with DOP of 266.9 and Mw of 43,250.
4. Conclusions
In this study cellulose nanofibrils were isolated from wheat
straw using alkaline steam explosion followed by chemical and
high shear treatment. The major constituent of these fibers was
found to be cellulose. TEM and AFM images confirmed fiber diameter of 30–50 nm. FT-IR and XRD studies evidenced about dissolution of lignin and hemicellulose with chemical treatment of the
fibers, which resulted in improved thermal stability. The cellulose
nanofibers were most stable to heat treatment followed by chemically treated fibers and untreated wheat straw fibers. Degree of
polymerization decreased from 2609.7 for steam exploded wheat
straw to 266.9 for cellulose nanofibers.
Acknowledgments
We gratefully acknowledge financial support rendered by All
India Council of Technical Education (AICTE), India and University
Grants Commission (UGC), India for the work.

References

1. Siqueira, G.; Bras, J.; Dufresne, A. Biomacromolecules 2009, 10, 425–432.
2. Soykeabkaew, N.; Supaphol, P.; Rujiravanit, R. Carbohydr. Polym. 2004, 58, 53–
63.
3. Okubo, K.; Fuji, T.; Thostenson, E. T. Composites Part A—Appl. S 2009, 40, 469–
475.
4. Abe, K.; Iwamoto, S.; Yano, H. Biomacromolecules 2007, 8, 3276–3278.
5. Ahola, S.; Salmi, J.; Johansson, L.-S.; Laine, J.; Ãsterberg, M. Biomacromolecules
2008, 9, 1273–1282.
6. Hitoshi, T.; Akira, A. Key Eng. Mat. 2007, 334–335, 389–392.
7. Said Azizi Samir, M. A.; Alloin, F.; Dufresne, A. Biomacromolecules 2005, 6, 612–
626.
8. Nishino, T.; Matsuda, I.; Hirao, K. Macromolecules 2004, 37, 7683–7687.
9. Fengel, D.; Wegner, G. Wood: Chemistry Ultrastructure Reactions; Walter de
Gruyter: Berlin, 1989.
10. Nakagaito, A.N.; Yano, H. American Chemical Society: Nanocomposites Based on
Cellulose Microfibril Cellulose Nanocomposites; Washington, DC, 2009; pp 151–
168.
11. Nakagaito, A. N.; Yano, H. Appl. Phys. A 2004, 78, 547–552.
12. Nakagaito, A. N.; Yano, H. Appl. Phys. A 2005, 80, 155–159.
13. Turbak, A. F.; Snyder, F. W.; Sandberg, K. R. J. Appl. Polym. Sci.: Appl. Polym.
Symp. 1983, 37, 813–827.
14. Andresen, M.; Johansson, L.; Tanem, B. S.; Stenius, P. Cellulose 2006, 13, 665–
677.
15. Stenstad, P.; Andresen, M.; Tanem, B. S.; Stenius, P. Cellulose 2008, 15, 35–45.
16. Liu, R.; Yu, H.; Huang, Y. Cellulose 2005, 12, 25–34.
17. Cherian, B. M.; Pothan, L. A.; Nguyen-Chung, T.; Mennig, G.; Kottaisamy, M.;
Thomas, S. J. Agric. Food Chem. 2008, 56, 5617–5627.
18. Purl, V. P.; Mamers, H. Biotech. Bioeng. 1983, 25, 3149–3162.
19. Battista, O. A.; Smith, P. A. Ind. Eng. Chem. 1962, 54, 20–29.
20. Structure of Cellulose and its Relation to Properties of Cellulose Fibers; Krassig, H.,

Ed.; Ellis Horwood: Chichester, U.K., 1985.
21. Segal, L.; Creely, J. J.; Martin, A. E., Jr.; Conrad, C. M. Text. Res. J. 1959, 29, 786–
794.
22. Bodor, G. Structural Investigation of Polymers; Ellis Horwood: New York, 1991.
23. Robert, E.; Adrian, F. A. W. J. Appl. Polym. Sci. 1989, 37, 2331–2340.
24. Yamashiki, T.; Matsui, T.; Saitoh, M.; Okajima, K.; Kamide, K.; Sawada, T. Brit.
Polym. J. 1990, 22, 12–128.
25. Xiao, B.; Sun, X. F.; Run, C. S. Polym. Degrad. Stab. 2001, 74, 307–319.
26. Sun, X. F.; Xu, F.; Sun, R. C.; Fowler, P.; Baird, M. S. Carbohydr. Res. 2005, 340,
97–106.
27. Sain, M.; Panthapulakkal, S. Ind. Crop. Prod. 2006, 23, 1–8.
28. Alemdar, A.; Sain, M. Bioresour. Technol. 2008, 99, 1664–1671.
29. Waleed, K. E.-Z.; Maha, M. I. Polym. Adv. Technol. 2003, 14, 623–631.
30. Gañán, P.; Cruz, J.; Garbizu, S.; Arbelaiz, A.; Mondragon, I. J. Appl. Polym. Sci.:
Appl. Polym. Symp. 2004, 94, 1489–1495.
31. Tang, L. C.; Hon, D. N. S.; Pan, S. H.; Zhu, Y. Q.; Wang, Z.; Wang, Z. Z. J. Appl.
Polym. Sci. 1996, 59, 483–488.
32. Ouajai, S.; Shanks, R. A. Polym. Degrad. Stabil. 2005, 89, 327–335.
33. Yang, P.; Kokot, S. J. Appl. Polym. Sci. 1996, 60, 1137–1146.
34. Yao, F.; Wu, Q.; Lei, Y.; Guo, W.; Xu, Y. Polym. Degrad. Stabil. 2008, 93, 90–98.
35. Antal, M. J.; Varhegyi, G. Ind. Eng. Chem. Res. 1995, 34, 703–717.
36. Gronli, M. G.; Varhegyi, G.; Blasi, C. D. Ind. Eng. Chem. Res. 2002, 41, 4201–4208.