Sugar Tech

DOI 10.1007/s12355-016-0507-1

RESEARCH ARTICLE

Characterization of Cellulose Nanocrystals Extracted

from Sugarcane Bagasse for Potential Biomedical Materials
Nga Tien Lam1 • Rungsima Chollakup2 • Wirasak Smitthipong3 •

Thidarat Nimchua4 • Prakit Sukyai1

Received: 7 October 2016 / Accepted: 26 December 2016

Ó Society for Sugar Research & Promotion 2017

Abstract Cellulose nanocrystals (CNCs) were extracted tively. A decrease in the thermal stability of CNCs was
by sulfuric acid from cellulose purified via an environ- examined by thermogravimetric analysis, and no evidence
mentally friendly method. In this study, cellulose obtained of cytotoxicity in the CNCs was obtained. The isolated
from sugarcane bagasse (SCB) using steam-exploded and CNCs from SCB may be considered as a potential
enzyme-treated pretreatment was confirmed using chemi- biomedical material.
cal composition analysis to have a 92.59 ± 0.12 whiteness
index and 87% a-cellulose content. The morphology of Keywords Cellulose nanocrystals  Cytotoxicity 
extracted CNCs, characterized using atomic force micro- Sugarcane bagasse  Xylanase
scopy images, transmission electron microscopy images
and energy-dispersive x-rays, showed the diameter and
length were in the ranges 9.8 ± 6.3 and 280.1 ± 73.3 nm, Introduction
respectively, with an expected ratio (L/d) of 20–25 and a
low concentration of sulfate (0.2%) on surface particles. In the last few decades, there has been a trend to utilize
Moreover, fourier transformed infrared spectroscopy and biodegradable and renewable material which is sourced
X-ray diffraction results demonstrated free noncellulosic from agricultural residues such as coconut husks (Rosa
contents and an improved crystallinity for CNCs, respec- et al. 2010) and sugarcane bagasse (Teixeira et al. 2011).
Sugarcane bagasse is the residual fraction of sugarcane
after juice extraction in the sugar production process
(Bhattacharya et al. 2008). Similar to other lignocellulose
components, SCB is consisted of a complex network of
& Prakit Sukyai carbohydrate polymers including lignin, hemicellulose,
[email protected] cellulose and pectin (Sun et al. 2004). A great deal of
1
Biotechnology of Biopolymer and Bioactive Compounds
previous literature indicated that nearly one-half (40–50%)
Laboratory, Department of Biotechnology, Faculty of Agro- of the raw material weight from SCB is cellulose (Bhat-
Industry, Kasetsart University, Chatuchak, Bangkok 10900, tacharya et al. 2008). Due to the high concentration of
Thailand cellulose, SCB becomes a potential source for cellulose
2
Kasetsart Agricultural and Agro-Industrial Product extraction to be used in recyclable materials. In recent
Improvement Institute (KAPI), Kasetsart University, decades, SCB has been well known for its three main
Chatuchak, Bangkok 10900, Thailand
components: (1) lignin, which is formed from complex
3
Department of Materials Science, Faculty of Science, aromatic rings (Zuluaga et al. 2009); (2) hemicellulose, an
Kasetsart University, Chatujak, Bangkok, Thailand
amorphous polymer with rich lateral chains consisting of
4
Enzyme Technology Laboratory, Bioresources Technology diverse sugars; and (3) cellulose identified by a linear
Unit, National Center for Genetic Engineering and
homopolymer made from anhydroglucose units linked by
Biotechnology (BIOTEC), 113 Thailand Science Park,
Phahonyothin Road, Khlong Nueng, Khlong Luang, b-glycosidic bonds (Gilbert and Kadla 2000). Generally,
Pathum Thani 12120, Thailand the disruption of the natural matrices of the lignin,

123
 Sugar Tech

hemicellulose, cellulose and others components has been Materials and Methods
carried out during using chemical agents and physical force
(Brinchi et al. 2013). Thermal pretreatment with steam- Sugarcane bagasse was obtained from the Kaset Thai
exploded force has been used in many studies as it is not International Sugar Corporation Public Company Limited
only acts as a defibrillating agent but also decomposes the factory, Nakhonsawan, Thailand. Xylanase (1568.57 ±
macro-polymers into low molecular weight fractions 2.31 U mL-1) was supplied by The National Center for
(Zheng et al. 2014). An environmentally friendly process Genetic Engineering and Biotechnology (BIOTEC), Thai-
with xylanase assistance to diminish the need for bleaching land Science Park, Thailand.
chemicals can be employed effectively after explosion
pretreatment (Saelee et al. 2016). This helps to partially Cellulose Extraction
dissolve the hemicellulose content and is followed by
sodium chloride to totally destroy noncellulosic matter Bagasse was washed with water to remove artifacts, fol-
(Zuluaga et al. 2009; Mandal and Chakrabarty 2011). lowed by drying at 55 °C for 24 h. Dried SCB was pre-
Cellulose in nano-sized particles—known as nanocel- treated using a steam-explosion machine (Nitto Koatsu 2.5
lulose (NC)—is a promising potential biomaterial, having a L, Japan) at a pressure of at least 1.3 MPa (190 °C) for
nanoscale dimension, high specific strength and modulus, 15 min. Then, the steam-exploded bagasse was digested
high surface area and unique optical properties (Dri et al. using xylanase, at 20 U g-1 (solid–liquid ratio 1:15 (w/
2013; Wu et al. 2012). A detailed study of cellulose fiber v)), under agitation. The experiment was carried out at
structure indicates that cellulose possesses high compact 50 °C for 2 h (Saelee et al. 2016). Bleached fiber was
domains, crystal areas, is stabilized via hydrogen bonding obtained after treating with 1.4% (w/v) sodium chlorite
and is disrupted by disordered fractions and amorphous (NaClO2) solution at pH 4 which was adjusted by acetic
regions. Thus, cellulose nanocrystals (CNCs) extraction is acid at 70 °C for 6 h (Mandal and Chakrabarty 2011).
the elimination of amorphous fractions in cellulose fiber by NaClO2 and acetic acid were added every hour until a
acid hydrolysis such as by using sulfuric acid (H2SO4) white bleached fiber was obtained. The resulting suspen-
(Mandal and Chakrabarty 2011) or hydrochloric acid (HCl) sion was filtered and washed with distilled water until the
(Teixeira et al. 2011). Typically, CNCs are rigid, rod-like pH reached 6–7.
crystals with a diameter in the range 10–20 nm and a
length of 100 nm (Mandal and Chakrabarty 2011; Teixeira Nanocellulose Preparation
et al. 2011). The main features that stimulate the use of
CNCs as polymer reinforcement agents are its large About 5 g of cellulose was dispersed in 100 mL of 60%
specific surface area, very high modulus of elasticity (ap- (w/v) sulfuric acid hydrolysis (solid–liquid ratio 1:20) at
proximately 150 GPa), biocompatibility and biodegrad- 45 °C for 75 min with continuous stirring (500 rpm). The
ability (Šturcová et al. 2005; Mariano et al. 2014; Petersson hydrolysis was quenched by adding 500 mL water to the
and Oksman 2006). In the mammalian body, cellulose reaction mixture, and then the slurry was washed with
exhibits a relative low protein adsorption and cell adhesion distilled water for 15 min at 15,000 rpm, using repeated
and a low immune response. Moreover, nanocellulose can centrifugation. The supernatant was removed from the
provide a cell-friendly environment to encourage cell sediment and replaced by clean distilled water and mixed;
attachment and proliferation as a special tissue bio-scaf- the centrifugation step continued until the pH of the
fold. No significant cytotoxicity to various human cell lines supernatant was 1. The last wash was conducted using
was found in previous studies (Zhou et al. 2013; Male et al. dialysis with distilled water until the wash water main-
2012). tained a constant pH of 7. Then, sonication (Elmasonic
There is no earlier report work on evaluating toxicity for Model S100H, Germany) was conducted in an ice bath for
the cell line of CNCs extracted from SCB. Moreover, this 15 min. The aqueous suspension was stored at 4 °C in a
study applied the process with xylanase-assisted pretreat- fridge for further experiments (Zhou 2012; Kumar et al.
ment for cellulose purification that has been recorded in 2014).
bleached agent reduction and also not discussed in any
previous literature for cellulose preparation to extract Color Reflectance Properties
CNCs. Thus, the objective of this study was to isolate
cellulose nanocrystals (CNCs) from extracted cellulose via The color values of fibers were measured using a col-
an environmentally friendly process and to characterize the orimeter (UltraScan XE, Hunter Associates Laboratory,
CNCs obtained in terms of their morphology, crystallinity, Reston, VA, USA). A white standard plate was used to
thermal behavior and cytotoxicity for biomedical calibrate the instrument with the color values of L = 100,
application. a = 0 and b = 0. The color of specimens was determined

123
Sugar Tech

using an ‘‘L’’ value range from 0 (black) to 100 (white), an Fourier Transform Infrared Spectrometer Analysis
‘‘a’’ value range from -80 (greenness) to 100 (redness) and
a ‘‘b’’ value range from -80 (blueness) to 70 (yellowness). Infrared spectra of the samples were recorded using FTIR
The total color difference (DE) and whiteness index (WI) spectra (Bruker Tensor 27 spectrometer, USA) at room
were calculated using the equations below (Rhim et al. temperature. The sample fractions were dried in an oven at
1999) 60 °C for 12 h, mixed with KBr and pressed into pellets.
 0:5 The samples were analyzed over the range 500–4000 cm–1.
DE ¼ DL2 þ Da2 þ Db2
h i0:5 X-Ray Diffraction
WI ¼ 100  ð100  LÞ2 þa2 þ b2
XRD measurements were taken using an X-ray diffrac-
where DL = Lstandard - Lsample; Da = astandard - asample;
tometer (Philips Analytical X’Pert, the Netherlands) using
Db = bstandard - bsample. Three measurements were taken
Cu-Ka radiation. The samples were scanned over a range
on each sample.
varying over 2h in the range 10–50°. The crystallinity
index (CrI) of cellulose materials was calculated from the
Chemical Composition height of the 200 peak (I200, 2h = 22.6°) and the intensity
minimum between the peaks at 200 and 110 (Iam,
Chemical composition of samples of untreated, steam-ex- 2h = 18°) using the Segal method and the equation below
ploded, enzyme-treated and bleached fibers was evaluated (Segal et al. 1962):
following a standard method of Technical Association of
I200  Iam
Pulp and Paper Industry standard (TAPPI). According to CrI% ¼  100
TAPPI, the lignin concentration was determined by T222 I200
om-98, followed by estimation of the holocellulose con- where I200 represents both a crystalline and an amorphous
tents (a-cellulose and hemicellulose) using the acid chlorite material, and Iam represents an amorphous material.
method. Afterward, the a-cellulose percentage was deter-
mined according to TAPPI T203 om-88, and hemicellulose Thermogravimetric Analysis
was determined by subtracting the a-cellulose component
from the holocellulose content. An average of three repli- The thermal properties of cellulose fibers were determined
cates was calculated for each sample. using a thermogravimetric analyzer (Mettler Toledo,
Model TGA/SDTA 851e, Switzerland) for TGA and
Scanning Electron Microscopy Each sample was placed derivative thermogravimetric analysis (DTG). The samples
on a steel cylinder and sputtered with gold. Surface mor- were heated from 50 to 600 °C with a scanning rate of
phologies were examined using a scanning electron 10 °C min-1 under a nitrogen atmosphere. The weight of
microscope (Hitachi model, Jeol JSM5600LV, Japan) at an sample was in range 2–5 mg.
accelerating voltage of 15–20 kV. The energy-dispersive
X-ray (EDX) diffractor attached with the SEM unit was Cytotoxicity
operated for CNCs elemental analysis.
Cytotoxicity analysis of the CNCs suspension was per-
Atomic Force Microscopy Measurements were taken in formed in accordance with ISO standard 10993-5 that eval-
tapping mode using an Asylum model MFP-3D AFM (Bio, uated the cellular response with the test specimen extracts by
USA) at ambient temperature. In the sample preparation, a 3-(4, 5-Dimethythiazolyl-2)-2, 5-diphenyltetrazolium bro-
drop of diluted CNCs aqueous suspension of colloid was mide (MTT) assay. For the indirect cytotoxicity test, mouse
dispersed on the surface of an optical glass substrate and fibroblast L929 cells (NCTC clone 929: CCL 1, American
allowed to dry at ambient temperature and analyzed Type Culture Collection [ATCC], of Strain L) were cultured
subsequently. in minimum essential medium (MEM) at a density of
1 9 105 cells mL-1 into a 96-well plate followed by incu-
Transmission Electron Microscopy A CNCs dilute bation for 24 ± 2 h (5 ± 1% CO2, 37 ± 1 °C and 95 ± 5%
aqueous suspension of 0.01% (w/v) was cast as one drop relative humidity). The CNCs sterilized by ethylene oxide
onto a copper grid with carbon coating followed by nega- gas (ETO) was placed in 96-well plates and extracted at
tive staining of uranyl acetate of 2% (w/v) for a few 37 ± 1 °C for 24 ± 2 h. The number of living cells after
minutes. Each sample was dried at room temperature and incubation was quantified using a 3-(4, 5-Dimethythiazolyl-
then observed at 100 keV using TEM analysis (Hitachi 2)-2, 5-diphenyltetrazolium bromide (MTT) assay. The
model HT7700, Japan). viability was calculated as a ratio between the blank and the

123
 Sugar Tech

sample absorbance that was measured using a microplate delignification usually results in lower color of fibers
reader (Bio-Tek, USA) at 570 nm. The lower the viability (Teixeira et al. 2010; Chen et al. 2011).
value, the higher the cytotoxic potential of the sample The chemical composition of the fibers was determined in
(Kirdponpattara et al. 2015; Kuzmenko et al. 2016). accordance with the TAPPI standards which were used for a-
cellulose (TAPPI T202 om-88) and lignin (TAPPI T222 om-
Morphology of Mouse Fibroblast L929 Cell Line 98). Raw fibers recorded a high cellulose concentration
(42.46 ± 0.04%) that combined with hemicellulose
The mouse fibroblast L929 cells (NCTC clone 929: CCL 1, (33.35 ± 0.48%) and lignin (20.97 ± 0.55%). However,
American Type Culture Collection [ATCC], of Strain L) steam-exploded fiber showed a decline in the hemicellulose
were seeded in MEM for 24 h. Then, 0.05 g mL-1 of the and lignin contents (14.44 ± 0.28% and 19.58 ± 0.01%,
CNCs extractive was used to replace the MEM, and a respectively). It is well documented that hemicellulose pre-
negative control with MEM was incubated for another 24 h sent on the surface has a weak structure that consists of short
at 37 °C. The influence of the CNCs on cells morphology lateral chains of different sugars (pentoses, hexoses and
was observed under a trinocular phase contrast microscope acetylated sugars), so that hemicellulose could be easily
(Leica DMIL, Germany). attacked by diluted acid, alkali or enzymes under mild con-
ditions (Sun et al. 2005). Similarly, hemicellulose was
destroyed under the high steam pressure which resulted in the
Result and Discussion disruption of the lignin and hemicellulose network and lignin
degradation to a certain extent (Zhang et al. 2008). Fur-
Color Reflectance Properties and Chemical thermore, due to its branched chains, hemicellulose was
Composition continuously degraded by xylanase. Under such circum-
stances, xylanase catalyzed deacetylated xylan substrates
It is well documented that the high concentration of non- that led to a slightly lower hemicellulose content
cellulosic components can be attributed to the pigment (12.73 ± 0.22%) of the enzyme-treated fiber than that of
content of fibers while white fibers have a greater amount steam-exploded fiber (14.44 ± 0.28%) (Roncero et al.
of cellulose; thus, the white color of the fiber also indicates 2003). At the end of the purification process, the a-cellulose
the successful removal of noncellulose matter (Teixeira content (87.4%) of the bleached fiber was twice that of the
et al. 2010). The whiteness index (WI) was used to estimate untreated fiber (42.46%). Moreover, the concentrations of
values related to the white color which was considered as a lignin and hemicellulose were substantially decreased after
measurement of change during the cellulose purification bleaching, with figures of 0.99 and 7.06%, respectively
process. The whiteness of untreated fibers was recorded (Table 2). In acidic media, sodium chlorite liberates chlorine
around 63.21 ± 1.33, which was brighter than for the that is widely used to increase the removal of lignin including
steam-exploded fibers (50.26 ± 1.5) and enzyme-treated aromatic components (Zuluaga et al. 2009). Bleached fibers
fibers (47.31 ± 0.97). Generally, the increased pigment in were successfully obtained while using an environmentally
the steam-exploded fibers resulted from generating lignin friendly methodology (Saelee et al. 2016). The decreases in
that contains chromophore and auxochrome. Hemicellulose the solid yield of each stage (Table 2) confirmed the removal
is also attributed to various sugar residues and branching of noncellulosic components (Kargarzadeh et al. 2012).
chemistry that may be ready to break and release at high
temperature in the steam process. The bleaching agent Morphology Analysis
showed its effect on obtaining bleached fiber (with 92.59
WI) after 6 h (Table 1; Fig. 1). The bleaching process may The SEM results showed that the morphology of the fiber
chemically remove colors from materials and leads to a obtained after each stage had noticeably changed. The
more uniform reflectance. Consequently, increased untreated fibers contained high concentration of

Table 1 Hunter color values (mean value ± standard deviation) of the L*, a* and b* parameters, the total color difference (E) and whiteness
index (WI) functions of the fibers at different stages of treatment
Sample L* a* b* E* WI*

Untreated fiber 66.17 ± 1.42 3.08 ± 0.09 14.13 ± 0.36 30.56 ± 1.29 63.21 ± 1.33
Steam-exploded fiber 52.55 ± 1.56 7.01 ± 0.24 13.15 ± 0.83 43.41 ± 1.49 50.26 ± 1.50
Enzyme-treated fiber 49.75 ± 1.02 7.13 ± 0.19 14.15 ± 0.19 46.35 ± 0.97 47.31 ± 0.97
Bleached fiber 94.79 ± 0.12 0.05 ± 0.01 05.27 ± 0.06 04.54 ± 0.05 92.59 ± 0.12

123
Sugar Tech

Fig. 1 Physical appearance of

untreated fiber (a), steam-
exploded fiber (b), enzyme-
treated fiber (c) and bleached
fiber (d)

Table 2 Chemical composition (mean value ± standard deviation) of sugarcane bagasse samples after cellulose purification
Sample Lignin (%) Hemicellulose (%) a-cellulose (%) Yield (%)

Untreated fiber 20.97 ± 0.55 33.35 ± 0.48 42.46 ± 0.04 100

Steam-exploded fiber 19.58 ± 0.01 14.44 ± 0.28 61.37 ± 0.43 77.40 ± 3.88
Enzyme-treated fiber 16.60 ± 0.02 12.73 ± 0.22 64.53 ± 0.73 68.32 ± 0.52
Bleached fiber 0.99 ± 0.06 7.07 ± 0.49 87.40 ± 0.19 50.21 ± 1.80

extractives (waxes, pectin, oil, etc.) on the fibers surface damage on the exterior area of fibers such as exfoliation,
which contributed to their superficial layers (Fig. 2a) caving in and fracturing due to the lack of a partial
(Mandal and Chakrabarty 2011; Kumar et al. 2012). fraction of xylans in the xylanase digestion (Fig. 2c)
Steam-exploded fibers had a smaller diameter than that of (Saelee et al. 2016). The bleaching process involving
untreated fibers and presented the elimination of surface treatment with sodium chlorite in an acid medium resulted
impurities along with defibrillation (Fig. 2b). During the in the lignin being eliminated via complex formation and
steam-explosion stage, the force of the steam acted as an depolymerization. Because of the existence of space on
agent to depolymerize and disintegrate the organization of the surface fiber, the bleaching agent readily accessed the
untreated fibers into fiber bundles, opened up the partic- interior of the structure. This action led to a diameter
ulate structure of the biomass and further removed decrease in the fibrils to approximately 10–20 lm, which
amorphous ingredients (lignin, hemicellulose, etc.) (Sun was smaller than the average size of the fiber bundles
and Cheng 2002; Abraham et al. 2011). This led to a size (75–150 lm) before the bleaching treatment. (Figure 2d)
reduction that enhanced the external surface area of the (Sun et al. 2004; Mandal and Chakrabarty 2011). More-
fibers for enzyme hydrolysis (Marzieh et al. 2015). over, the bleached fibers were observed as round, fine
Clearly, the effect of the enzyme treatment exposed some bundles of fibers with clean, smooth surfaces that

123
 Sugar Tech

Fig. 2 SEM images of untreated fiber (a), steam-exploded fiber (b), enzyme-treated fiber (c) and bleached fiber (d)

confirmed the removal of wax, pectin, lignin and hemi- EDX was used to detect the elemental components in
cellulose (Kumar et al. 2012). samples to determine sulfuric acid removal by the dialysis
process in the CNCs extraction process. Figure 4 shows the
Atomic Force Microscopy Analysis, Transmission EDX spectrum of extracted cellulose nanocrystals and reveals
Electron Microscopy and Energy-Dispersive X-Ray the peaks for carbon, oxygen and sulfur related to their binding
Diffraction energies, respectively. The CNCs spectrum contained 0.2%
elemental impurity of sulfur and main components of 52.71%
The morphological properties of the CNCs suspension can for carbon and 47.09% for oxygen (Table 3). The remaining
be observed in Fig. 3a. The CNCs colloid verified the well- sulfate groups in the CNCs at a low percentage confirmed that
suspended state of nanoparticles in distilled water. AFM the CNCs colloid was acid-free (Ranby 1949).
detected the needle-like morphology of the extracted cel-
lulose at a nanoscale which was similar to the CNCs Fourier Transform Infrared Spectroscopy
obtained by Kim and Song (2015). Both the TEM (Fig. 3b)
and AFM (Fig. 3c) images display some aggregated FTIR spectroscopy was carried out to study the chemical
nanoparticles and individual nanocrystals perhaps due to functional group changes during the process. Figure 5
the high specific surface area and strong interaction bonds shows the FTIR spectra of untreated fiber, steam-exploded
presented among CNCs. From Fig. 3a, b, the CNCs are fiber, enzyme-treated fiber, bleached fiber and CNCs. The
uniformly dispersed in width and exhibit irregular length. FTIR spectra of all samples exhibited a broad band in the
The various dimensions of CNCs can be explained by the region 3500–3200 cm-1 that corresponded to the O–H
differences in raw material sources and extraction condi- stretching vibration of the OH groups in cellulose mole-
tions (Eichhorn et al. 2010). The average width of CNCs cules (Mandal and Chakrabarty 2011; Elanthikkal et al.
was found to be around 9.8 ± 6.3 nm, and the average 2010). Moreover, the characteristic C–H stretching vibra-
length was 280.1 ± 73.3 nm, which resulted in an aspect tion around 2894 cm-1 was found in all samples (Khalil
ratio (L/d) of around 20–25 (Fig. 3d). et al. 2001). The peak at 1737 cm-1 was associated with

123
Sugar Tech

Fig. 3 Physical appearance of

cellulose nanocrystals (a), TEM
image (b), AFM image (c) and
height distribution (d)

Table 3 Element component (mean value ± standard deviation) of

cellulose nanocrystals
Elements Weight (%) Atomic (%)

CK 45.45 ± 0.31 52.71

OK 54.05 ± 0.31 47.09
SK 0.47 ± 0.03 0.20
Total 100

1606 cm-1 were related to the aromatic –C=C– stretch of

the aromatic ring present in lignin (Wang et al. 2010). The
peak at 1425 cm-1 was due to the –CH2– bending (Elan-
Fig. 4 Energy-dispersive X-ray diffraction result of cellulose
nanocrystals thikkal et al. 2010). The vibration peak detected at 1473
and 1365 cm-1 in all samples has been related to the
bending vibration of the C–H symmetric and asymmetric
the C=O stretching vibration either of the acetyl and uronic deformations, respectively (Sun et al. 2005). The absor-
ester groups of the hemicelluloses or of the ester linkage of bance at 1327 cm-1 is attributed to the C–C and C–O
carboxylic group of the ferulic and p-coumaric acids of skeletal vibrations (Sun et al. 2005). The peak at
lignin and/or hemicelluloses (Li et al. 2014; Sun et al. 1245 cm-1 presents the C–O out-of-plane stretching
2005; Elanthikkal et al. 2010). The band at 1632 cm-1 was vibration of the aryl group in lignin (Troedec et al. 2008).
due to the bending mode of the absorbed water (Sun et al. The peak observed in the spectrum at 1054 cm-1 is due to
2005; Troedec et al. 2008). The peaks at 1514 and the C–O–C pyranose ring (antisymmetric in phase ring)

123
 Sugar Tech

Fig. 5 FTIR spectra of untreated fiber (a), steam-exploded fiber (b), enzyme-treated fiber (c), bleached fiber (d) and cellulose nanocrystals (e)

stretching vibration (Mandal and Chakrabarty 2011). The bleached fiber spectrum (Rosa et al. 2010). These observa-
897 cm-1 band in all samples represented the glycosidic – tions confirmed the successful elimination of lignin in the
C–H deformation, with a ring vibration contribution and – bleached fibers. There was no difference between the spec-
O–H bending which are known characteristics of b-gly- trum of nanocellulose and the bleached fibers.
cosidic linkages between the anhydroglucose units in cel-
lulose (Alemdar and Sain 2008). X-Ray Diffraction
The shape of the peak at 1730 cm-1 found in the spectrum
of untreated fibers (Fig. 5a) was associated with the C–O To evaluate the effect of each treatment on the resulting
stretching vibration for the acetyl and ester linkage in lignin material based on the changing in crystallinity of the fibers,
or hemicellulose. Therefore, steam-exploded fiber and XRD was carried out. Figure 6 presents the XRD patterns
enzyme-treated fiber did not have this peak due to the of fibers obtained at different stages of purification. The
removal of the hemicellulose and lignin to a certain extent crystallinity percentage of fibers at each stage is illustrated
(Saelee et al. 2016). The intensity of peak at 1249 cm-1 in Table 4.
stood for the C–O out-of-plane stretching vibration of the From Fig. 6, all of the samples recorded similar
aryl group in lignin and was substantially reduced in the diffraction patterns. However, the difference of slight
spectrum of bleached fiber obtained after sodium chloride intensity changes in the peaks was notable, as it repre-
treatment (Troedec et al. 2008). Moreover, the spectral band sented minor changes in the level of order in the samples
at 1604, 1515 and 1460 cm-1 in the untreated fiber curve (Elanthikkal et al. 2010). The XRD pattern of each kind of
represented aromatic ring vibration and C–H deformation fiber showed peaks around 2h = 16° and 22.5° assigned to
vibration of lignin. These peaks were also absent in the plane (110) and (200), which were believed to represent the

123
Sugar Tech

Fig. 6 XRD of untreated fiber

(a), steam-exploded fiber (b),
enzyme-treated fiber (c),
bleached fiber (d) and cellulose
nanocrystals (e)

Table 4 Crystallinity Index (CrI) (mean value ± standard deviation)

amorphous domains that were readily degraded during acid
of the fibers after different stages of treatment hydrolysis (Azizi Samir et al. 2004).
Sample Crystallinity Index (%CrI)
Thermal Analysis
Untreated fiber 47.38 ± 1.18
Steam-exploded fiber 49.36 ± 0.81 Figure 7a, b shows the thermal properties resulting from
Enzyme-treated fiber 51.23 ± 0.98 the including thermogravimetry (TG) and derivative ther-
Bleached fiber 56.06 ± 1.01 mogravimetry (DTG) analysis of fibers. The TG graph
Cellulose nanocrystals 68.54 ± 1.30 indicates changes in weight during heating, and their
derivatives show changes in the TG slope which may not
be obvious from the TG curve. Table 5 displays the onset
polymorphs of cellulose I (Klemm et al. 2005). An of thermal decomposition (Ton) indicating the beginning of
increased trend toward crystallinity after each treatment degradation temperature and the temperature of maximum
stage confirmed the hemicellulose and lignin removal, with decomposition (Tmax) referred to as the temperature of the
figures of 49.36 ± 0.81% for steam-exploded fiber and maximum rate of degradation. Clearly, the initial mass
51.23 ± 0.98% for enzyme-treated fiber. The bleached losses starting at 75 °C for untreated, steam-exploded,
fiber recorded 56.06 ± 1.01% of CrI, which was higher enzyme-treated, bleached fibers and CNCs indicated the
than for untreated fiber (47.38 ± 1.18%) due to the elim- evaporation of loose, surface-bound moisture (H2O). The
ination of noncellulosic components (Saelee et al. 2016). It intermolecular H-bonded water was evaporated at around
is well documented that the cellulose structure includes a 100 °C in all samples. The degradation of untreated fiber
strong connection between the amorphous domain and the started at 280 °C, and the rate of degradation reached its
crystalline region by hydrogen bonding, which results in an peak at 357 °C (observed from the DTG curve) while that
ordered system with crystal-like properties. (Alemdar and of bleached fibers occurred at 328 °C and achieved the
Sain 2008). The effect of sulfuric acid in the crystallinity maximum rate degradation at 351 °C. The degradation of
improvement of CNCs (68.54 ± 1.30%) in comparison CNCs occurred afterward at 250 °C, showing an additional
with bleached fiber is shown in Fig. 6e and could be small shoulder, and the rate of degradation was reached at
explained by the CNCs structure being regarded as free 268 °C (also observed from the DTG curve). The untreated

123
 Sugar Tech

(a)

(b)

Fig. 7 TG (a) and DTG (b) curves of untreated fiber, steam-exploded fiber, enzyme-treated fiber, bleached fiber and cellulose nanocrystals

fiber illustrated a shoulder in its DTG curve at approxi- respectively. The same trend was observed in the enzyme-
mately 300 °C that resembled hemicellulose decomposi- treated fiber and bleached fiber. The bleaching process was
tion. In the curve of the steam-exploded fiber, the lack of pyrolyzed at a slightly lower temperature of 328 °C
this shoulder peak confirmed hemicellulose elimination. (weight loss 10%) compared to that of xylanase-treated
Moreover, the increase in Ton in steam-exploded fibers fiber and reached a Tmax of 351 °C (weight loss 61%) due
revealed the effect of the steam-explosion treatment on the to diameter of bleached fibers reduction. The improvement
removal of hemicellulose and partial lignin degradation. In in thermal stability could have been due to the noncellu-
the second stage, the removal of hemicellulose by xylanase losic components, which were considered low crystal
treatment resulted in the improved thermal stability of the structures, being liberated during the bleaching process.
enzyme-treated fibers. A DTG analysis of the enzyme- With the CNCs, thermal degradation occurred in a lower
treated fibers showed that Ton and Tmax shifted to 332 °C temperature range starting at a Ton of 250 °C and extending
(weight loss 6.2%) and 357 °C (weight loss 45.28%), to a Tmax of 268 °C. Simultaneously, the CNCs were less

123
Sugar Tech

Table 5 Thermal properties of the fibers after different stages of treatment

Sample Tonset (°C) Mass loss (%) Tmax (°C) Mass loss (%) Residue at 600 °C (%)

Untreated fiber 280 07.00 357 58.30 15.63

Steam-exploded fiber 330 07.80 359 49.41 19.62
Enzyme-treated fiber 332 06.20 357 45.28 22.16
Bleached fiber 328 10.00 351 61.03 08.16
Cellulose nanocrystals 250 03.20 268 26.21 07.25

thermally stable than bleached fibers, as the DTG curve of perhaps because the acid-hydrolyzed cellulose was reduced
the CNCs displays three stages of pyrolysis, mainly at to the size of nanometers and this increased the surface area
240–300, 300–380 and 450–550 °C, respectively (Wang and the number of free ends in the chain that were more
et al. 2007). The Tmax presented as the main peak in the accessible to degradation. Furthermore, the CNCs illus-
DTG curve was at a much lower temperature than that of trated two humps or shoulders in close proximity on the
native cellulose (around 400 °C) (George et al. 2008) DTG curve. The lower temperature degradation was

Fig. 8 Optical microscopy

images (9100) of cultured L929
cells after one day of incubation
a negative control, c CNCs,
e positive control and after MTT
assay of b negative control,
d CNCs and f positive control

123
 Sugar Tech

crystallinity percentage. However, the TGA curve illus-

trated a decline in the thermal persistence of nanoparticles
due to the introduction of sulfate groups on their surfaces.
Furthermore, MTT assay revealed the potential of CNCs
for biomedical application, with a cell viability of 83%.

Acknowledgements The authors are grateful to the Thailand

Research Fund (TRF) for financial support, the Department of
Biotechnology, Faculty of Agro-Industry and Kasetsart University,
Thailand for supplying facilities, the Scholarship Program for Inter-
national Graduate Students 2014; and the National Center for Genetic
Engineering and Biotechnology (BIOTEC), Thailand for providing
Fig. 9 Cell viability results (mean value) of CNCs, negative and xylanase enzyme.
positive control (bars represent standard deviation of the mean)
Funding This study was funded by Thailand Research fund (Grant
Number TRF5850012).
associated with the sulfated region of nanoparticles during
pyrolysis while the other was due to degradation of the un-
sulfated part (Roman and Winter 2004). Compliance with Ethical Standards

Conflict of interest All authors of this research paper declare that

Cytotoxicity that have no conflict of interest.

Cytotoxicity analysis of the CNCs used in this study was

performed in accordance with ISO standard 10993-5:2009 References
(E) Annex C. In this study, the MTT assay was performed
to evaluate the cell proliferation of the L929 cell line. The Abraham, E., B. Deepa, L.A. Pothan, M. Jacob, S. Thomas, U.
microscopic morphology of the L929 cell line that was Cvelbar, and R. Anandjiwala. 2011. Extraction of nanocellulose
fibrils from lignocellulosic fibres: A novel approach. Carbohy-
observed following culture in positive (cytotoxic), negative drate Polymers 86 (4): 1468–1475. doi:
(noncytotoxic) and CNCs extractive media is illustrated in 10.1016/j.carbpol.2011.06.034.
Fig. 8. There was no difference in the L929 cell mor- Alemdar, Ayse, and Mohini Sain. 2008. Biocomposites from wheat
phology after incubating for 24 h between the negative straw nanofibers: Morphology, thermal and mechanical proper-
ties. Composites Science and Technology 68 (2): 557–565. doi:
control and the media with the presence of CNCs. Figure 9 10.1016/j.compscitech.2007.05.044.
shows the cell viability results comparison of CNCs with Bhattacharya, Deepanjan, Louis T. Germinario, and William T.
the negative and positive controls. The viability threshold Winter. 2008. Isolation, preparation and characterization of
for the cytotoxicity analysis was 70%. The relative cell cellulose microfibers obtained from bagasse. Carbohydrate
Polymers 73 (3): 371–377. doi:10.1016/j.carbpol.2007.12.005.
numbers cultured with the extraction medium including Brinchi, L., F. Cotana, E. Fortunati, and J.M. Kenny. 2013.
CNCs was significantly higher than that of the positive Production of nanocrystalline cellulose from lignocellulosic
control plate, with figures of 83 and 0%, respectively. biomass: Technology and applications. Carbohydrate Polymers
CNCs have been well recognized as nontoxic materials for 94 (1): 154–169. doi:10.1016/j.carbpol.2013.01.033.
Chen, Wenshuai, Yu. Haipeng, Yixing Liu, Peng Chen, Mingxin
cells showing more than 70% cell viability (Dong et al. Zhang, and Yunfei Hai. 2011. Individualization of cellulose
2012). This result suggested that the extract from CNCs nanofibers from wood using high-intensity ultrasonication com-
produced no harmful effects to cell growth and thus, it can bined with chemical pretreatments. Carbohydrate Polymers 83
be concluded that CNCs are a material with no cytotoxic (4): 1804–1811. doi:10.1016/j.carbpol.2010.10.040.
Dong, S., A.A. Hirani, K.R. Colacino, Y.W. Lee, and M. Roman.
potential. 2012. Cytotoxicity and cellular uptake of cellulose nanocrystals.
Nano Life 02 (03): 124106. doi:10.1142/S1793984412410061.
Dri, Fernando L., Louis G. Hector, Robert J. Moon, and Pablo D.
Conclusion Zavattieri. 2013. Anisotropy of the elastic properties of crys-
talline cellulose Ib from first principles density functional theory
with Van Der Waals interactions. Cellulose 20 (6): 2703–2718.
Cellulose nanocrystals (CNCs) were successfully obtained doi:10.1007/s10570-013-0071-8.
using acid hydrolysis from bleached fibers that could be Eichhorn, S.J., A. Dufresne, M. Aranguren, N.E. Marcovich, J.R.
purified by steam-explosion and enzyme-assisted pretreat- Capadona, S.J. Rowan, C. Weder, et al. 2010. Review: Current
international research into cellulose nanofibres and nanocom-
ment. The morphological properties indicated a needle-like posites. Journal of Materials Science 45 (1): 1–33. doi:
shape of CNCs with an aspect ratio (L/d) of 20–25. The 10.1007/s10853-009-3874-0.
results from FTIR spectroscopy and XRD also confirmed Elanthikkal, Silviya, Unnikrishnan Gopalakrishnapanicker, Soney
noncellulose removal and an improvement in the Varghese, and James T. Guthrie. 2010. Cellulose microfibres

123
Sugar Tech

produced from banana plant wastes: Isolation and characteriza- Carbohydrate Polymers 86 (3): 1291–1299. doi:
tion. Carbohydrate Polymers 80 (3): 852–859. doi: 10.1016/j.carbpol.2011.06.030.
10.1016/j.carbpol.2009.12.043. Mariano, Marcos, Nadia El Kissi, and Alain Dufresne. 2014.
George, Johnsy, V.A. Sajeevkumar, R. Kumar, K.V. Ramana, S.N. Cellulose nanocrystals and related nanocomposites: Review of
Sabapathy, and A.S. Bawa. 2008. Enhancement of thermal some properties and challenges. Journal of Polymer Science Part
stability associated with the chemical treatment of bacterial B: Polymer Physics 52 (12): 791–806. doi:10.1002/polb.23490.
(Gluconacetobacter xylinus) cellulose. Journal of Applied Poly- Marzieh, Shafiei, Kumar Rajeev, and Keikhosro Karimi. 2015.
mer Science 108 (3): 1845–1851. doi:10.1002/app.27802. Pretreatment of lignocellulosic biomass. In Lignocellulose-based
Gilbert, R.D., and J.F. Kadla. 2000. Polysaccharides-cellulose. In bioproducts, vol. 1, ed. K. Karimi, 85–154. Switzerland:
Biopolymers from renewable resources. Macromolecular sys- Springer International Publishing.
tems—material approach, ed. D. Kaplan, 47–90. Berlin, Heidel- Petersson, L., and K. Oksman. 2006. Biopolymer based nanocom-
berg: Springer. posites: Comparing layered silicates and microcrystalline cellu-
Kargarzadeh, Hanieh, Ishak Ahmad, Ibrahim Abdullah, Alain lose as nanoreinforcement. Composites Science and Technology
Dufresne, Siti Yasmine Zainudin, and Rasha M. Sheltami. 66 (13): 2187–2196. doi:10.1016/j.compscitech.2005.12.010.
2012. Effects of hydrolysis conditions on the morphology, Ranby, B.G. 1949. Aqueous colloidal solutions of cellulose micelles.
crystallinity, and thermal stability of cellulose nanocrystals Acta Chemica Scandinavica 3: 649–650.
extracted from kenaf bast fibers. Cellulose 19 (3): 855–866. doi: Rhim, J.W., Y. Wu, C.L. Weller, and M. Schnepf. 1999. Physical
10.1007/s10570-012-9684-6. characteristics of a composite film of soy protein isolate and
Khalil, H.P.S.A., H. Ismail, H.D. Rozman, and M.N. Ahmad. 2001. propyleneglycol alginate. Journal of Food Science 64 (1):
The effect of acetylation on interfacial shear strength between 149–152. doi:10.1111/j.1365-2621.1999.tb09880.x.
plant fibres and various matrices. European Polymer Journal 37 Roman, Maren, and William T. Winter. 2004. Effect of sulfate groups
(5): 1037–1045. doi:10.1016/s0014-3057(00)00199-3. from sulfuric acid hydrolysis on the thermal degradation
Kim, D.H., and Y.S. Song. 2015. Anisotropic optical film embedded behavior of bacterial cellulose. Biomacromolecules 5 (5):
with cellulose nanowhisker. Carbohydrate Polymers 130: 1671–1677. doi:10.1021/bm034519?.
448–454. doi:10.1016/j.carbpol.2015.05.033. Roncero, M.Blanca, Antonio L. Torres, Jos F. Colom, and Teresa Vidal.
Kirdponpattara, S., A. Khamkeaw, N. Sanchavanakit, P. Pavasant, 2003. Effect of xylanase on ozone bleaching kinetics and properties
and M. Phisalaphong. 2015. Structural modification and charac- of Eucalyptus kraft pulp. Journal of Chemical Technology and
terization of bacterial cellulose-alginate composite scaffolds for Biotechnology 78 (10): 1023–1031. doi:10.1002/jctb.893.
tissue engineering. Carbohydrate Polymers 132: 146–155. doi: Rosa, M.F., E.S. Medeiros, J.A. Malmonge, K.S. Gregorski, D.F.
10.1016/j.carbpol.2015.06.059. Wood, L.H.C. Mattoso, G. Glenn, W.J. Orts, and S.H. Imam.
Klemm, Dieter, Brigitte Heublein, Hans-Peter Fink, and Andreas 2010. Cellulose nanowhiskers from coconut husk fibers: Effect
Bohn. 2005. Cellulose: Fascinating biopolymer and sustainable of preparation conditions on their thermal and morphological
raw material. Angewandte Chemie International Edition 44 (22): behavior. Carbohydrate Polymers 81 (1): 83–92. doi:
3358–3393. doi:10.1002/anie.200460587. 10.1016/j.carbpol.2010.01.059.
Kumar, Anuj, Yuvraj Singh Negi, Nishi Kant Bhardwaj, and Veena Saelee, Kullasatri, Naiyasit Yingkamhaeng, Thidarat Nimchua, and
Choudhary. 2012. Synthesis and characterization of methylcel- Prakit Sukyai. 2016. An environmentally friendly xylanase-assisted
lulose/PVA based porous composite. Carbohydrate Polymers 88 pretreatment for cellulose nanofibrils isolation from sugarcane
(4): 1364–1372. doi:10.1016/j.carbpol.2012.02.019. bagasse by high-pressure homogenization. Industrial Crops and
Kumar, Anuj, YuvrajSingh Negi, Veena Choudhary, and NishiKant Products 82: 149–160. doi:10.1016/j.indcrop.2015.11.064.
Bhardwaj. 2014. Microstructural and mechanical properties of Samir, Azizi, My Ahmed Said, Fannie Alloin, Jean-Yves Sanchez,
porous biocomposite scaffolds based on polyvinyl alcohol. and Alain Dufresne. 2004. Cellulose nanocrystals reinforced
Nano-Hydroxyapatite and Cellulose Nanocrystals. Cellulose 21 poly(oxyethylene). Polymer 45 (12): 4149–4157. doi:
(5): 3409–3426. doi:10.1007/s10570-014-0339-7. 10.1016/j.polymer.2004.03.094.
Kuzmenko, Volodymyr, Theodoros Kalogeropoulos, Johannes Thun- Segal, L., J.J. Creely, A.E. Jr Martin, and C.M. Conrad. 1962. An
berg, Sara Johannesson, Daniel Hägg, Peter Enoksson, and Paul empirical method for estimating the degree of crystallinity of
Gatenholm. 2016. Enhanced growth of neural networks on native cellulose using the x-ray diffractometer. Textile Research
conductive cellulose-derived nanofibrous scaffolds. Materials Journal 29: 786–794.
Science and Engineering, C: Materials for Biological Applica- Šturcová, Adriana, Geoffrey R. Davies, and Stephen J. Eichhorn.
tions 58: 14–23. doi:10.1016/j.msec.2015.08.012. 2005. Elastic modulus and stress-transfer properties of tunicate
Le Troedec, Marianne, David Sedan, Claire Peyratout, Jean Pierre cellulose whiskers. Biomacromolecules 6 (2): 1055–1061. doi:
Bonnet, Agnès Smith, René Guinebretiere, Vincent Gloaguen, 10.1021/bm049291k.
and Pierre Krausz. 2008. Influence of various chemical treat- Sun, Ye, and Jiayang Cheng. 2002. Hydrolysis of lignocellulosic
ments on the composition and structure of hemp fibres. materials for ethanol production: A review. Bioresource Tech-
Composites Part A Applied Science and Manufacturing 39 (3): nology 83 (1): 1–11.
514–522. doi:10.1016/j.compositesa.2007.12.001. Sun, J.X., X.F. Sun, R.C. Sun, and Y.Q. Su. 2004. Fractional
Li, M., L.J. Wang, D. Li, Y.L. Cheng, and B. Adhikari. 2014. extraction and structural characterization of sugarcane bagasse
Preparation and characterization of cellulose nanofibers from de- hemicelluloses. Carbohydrate Polymers 56 (2): 195–204. doi:
pectinated sugar beet pulp. Carbohydrate Polymers 102: 10.1016/j.carbpol.2004.02.002.
136–143. doi:10.1016/j.carbpol.2013.11.021. Sun, X.F., F. Xu, R.C. Sun, P. Fowler, and M.S. Baird. 2005.
Male, Keith B., Alfred C.W. Leung, Johnny Montes, Amine Kamen, Characteristics of degraded cellulose obtained from steam-
and John H.T. Luong. 2012. Probing inhibitory effects of exploded wheat straw. Carbohydrate Research 340 (1): 97–106.
nanocrystalline cellulose: Inhibition versus surface charge. doi:10.1016/j.carres.2004.10.022.
Nanoscale 4 (4): 1373–1379. doi:10.1039/C2NR11886F. Teixeira, Eliangela de Morais, Ana Carolina Corrêa, Alexandra
Mandal, A., and D. Chakrabarty. 2011. Isolation of nanocellulose Manzoli, Fabio de Lima Leite, Cauê Ribeiro de Oliveira, and
from waste sugarcane bagasse (scb) and its characterization. Luiz Henrique Capparelli Mattoso. 2010. Cellulose nanofibers

123
 Sugar Tech

from white and naturally colored cotton fibers. Cellulose 17 (3): lignin in wheat straw. Bioresource Technology 99 (17):
595–606. doi:10.1007/s10570-010-9403-0. 8512–8515. doi:10.1016/j.biortech.2008.03.028.
Teixeira, Eliangela, Thalita de Morais, Jessika Bondancia, Kelcilene Zheng, Yi, Jia Zhao, Xu Fuqing, and Yebo Li. 2014. Pretreatment of
Bruna Ricardo Teodoro, Ana Carolina Corrêa, José Manoel lignocellulosic biomass for enhanced biogas production. Pro-
Marconcini, and Luiz Henrique Caparelli Mattoso. 2011. gress in Energy and Combustion Science 42: 35–53. doi:
Sugarcane bagasse whiskers: Extraction and characterizations. 10.1016/j.pecs.2014.01.001.
Industrial Crops and Products 33 (1): 63–66. doi: Zhou, Y.M. 2012. Effect of nanocellulose isolation techniques on the
10.1016/j.indcrop.2010.08.009. formation of reinforced poly(vinyl alcohol) nanocomposite
Wang, Neng, Enyong Ding, and Rongshi Cheng. 2007. Thermal films. Express Polymer Letters 6 (10): 794–804. doi:
degradation behaviors of spherical cellulose nanocrystals with 10.3144/expresspolymlett.2012.85.
sulfate groups. Polymer 48 (12): 3486–3493. doi: Zhou, Chengjun, Qingfeng Shi, Weihong Guo, Lekeith Terrell,
10.1016/j.polymer.2007.03.062. Ammar T. Qureshi, Daniel J. Hayes, and Wu Qinglin. 2013.
Wang, Wei-Ming, Zai-Sheng Cai, Yu. Jian-Yong, and Zhao-Peng Electrospun bio-nanocomposite scaffolds for bone tissue engi-
Xia. 2010. Changes in composition, structure, and properties of neering by cellulose nanocrystals reinforcing maleic anhydride
jute fibers after chemical treatments. Fibers and Polymers 10 (6): grafted PLA. ACS Applied Materials and Interfaces 5 (9):
776–780. doi:10.1007/s12221-009-0776-3. 3847–3854. doi:10.1021/am4005072.
Wu, Xiawa, Robert J. Moon, and Ashlie Martini. 2012. Crystalline Zuluaga, Robin, Jean Luc Putaux, Javier Cruz, Juan Vélez, Iñaki
cellulose elastic modulus predicted by atomistic models of Mondragon, and Piedad Gañán. 2009. Cellulose microfibrils
uniform deformation and nanoscale indentation. Cellulose 20 from banana rachis: Effect of alkaline treatments on structural
(1): 43–55. doi:10.1007/s10570-012-9823-0. and morphological features. Carbohydrate Polymers 76 (1):
Zhang, L.H., D. Li, L.J. Wang, T.P. Wang, L. Zhang, X.D. Chen, and 51–59. doi:10.1016/j.carbpol.2008.09.024.
Z.H. Mao. 2008. Effect of steam explosion on biodegradation of

123