International Journal of Biological Macromolecules 101 (2017) 553–561

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

journal homepage: www.elsevier.com/locate/ijbiomac

Cellulose nanocrystals from acacia bark–Influence of solvent

extraction
Ticiane Taflick a,b,∗ , Luana A. Schwendler a , Simone M.L. Rosa a , Clara I.D. Bica a ,
Sônia M.B. Nachtigall a
a
Chemistry Institute, Federal University of Rio Grande do Sul/UFRGS, 9500 Bento Gonçalves Ave., Porto Alegre, 91501-970, Brazil
b
Sul-rio-grandense Federal Institute for Education, Science and Technology, 100 Copacabana Ave., Sapucaia do Sul, 93216-120, Brazil

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

Article history: The isolation of cellulose nanocrystals from different lignocellulosic materials has shown increased inter-
Received 15 July 2016 est in academic and technological research. These materials have excellent mechanical properties and
Received in revised form can be used as nanofillers for polymer composites as well as transparent films for various applications.
13 September 2016
In this work, cellulose isolation was performed following an environmental friendly procedure without
Accepted 14 March 2017
chlorine. Cellulose nanocrystals were isolated from the exhausted acacia bark (after the industrial pro-
Available online 18 March 2017
cess of extracting tannin) with the objective of evaluating the effect of the solvent extraction steps on the
characteristics of cellulose and cellulose nanocrystals. It was also assessed the effect of acid hydrolysis
Keywords:
Cellulose nanocrystals
time on the thermal stability, morphology and size of the nanocrystals, through TGA, TEM and light scat-
Acacia bark tering analyses. It was concluded that the extraction step with solvents was important in the isolation of
Solvent extraction cellulose, but irrelevant in the isolation of cellulose nanocrystals. Light scattering experiments indicated
Light scattering that 30 min of hydrolysis was long enough for the isolation of cellulose nanocrystals.
Morphology © 2017 Elsevier B.V. All rights reserved.
Thermal properties

1. Introduction nolics, simple sugars, pectins, mucilages, gums, terpenes, starch,

In recent years, there has been a growing interest in the
study of sustainable raw materials which present potential use
in nanotechnology. In this sense, cellulose nanocrystals (CNC)
derived from plant fibers have attracted interest as new materi-
als mainly for polymer reinforcement. The major components of
plant fibers are cellulose (40–50%), hemicellulose (20–30%) and
lignin (10–18%) [1]. Besides these components, variable amounts
of solvent extractable compounds and inorganic molecules are also
found.
Obtaining cellulose nanocrystals from various types of ligno-
cellulosic arrays involves a series of processes that may start with
solvent extraction steps using solvents of different polarities or sol-
vent mixtures to remove extractives [2–4]. This is a high cost and
time-consuming step. Chemically, extractives in wood are com-
ponents of low molecular weight. They include a wide range of
substances such as flavonoids, lignans, stilbenes, tannins, inorganic
salts, fats, waxes, alkaloids, proteins, simple and complex phe-
∗ Corresponding author at: Sul-rio-grandense Federal Institute for Education, Sci-
ence and Technology, 100 Copacabana Ave., Sapucaia do Sul, 93216-120, Brazil.
E-mail address: ticitaflick22@gmail.com (T. Taflick).
glycosides, saponins, and essential oils [5]. Water-soluble wood
extractives mainly consist of organic salts, sugars, some polysac-
charides and phenolic substances [6]. The composition and amount
of the extractives depend on factors such as wood species, wood age
and the location of the wood in the tree [5]. Bark is the part of the
plant where the soluble substances are mainly present [6].
Following solvent extraction, lignin and hemicellulose are selec-
tively removed from the fiber in the pulping step [7]. The most
commonly used process involves heating with alkaline solutions.
In general, after this step, the pulp still shows dark color and needs
the use of bleaching processes to achieve greater level of brightness
[8]. Conventional bleaching processes involve the use of chemicals
based on chlorine (chlorine, chlorine dioxide, sodium hypochlo-
rite), usually in a series of steps depending on the desired degree of
whiteness. The major drawback of these processes is the formation
of toxic organic compounds, mainly dioxins. Because of this, new
Totally Chlorine Free (TCF) bleaching sequences have been devel-
oped, which are mainly based on oxygen, hydrogen peroxide and
ozone [7].
Once cellulose is isolated from the plant, CNC can be obtained
by acid hydrolysis under controlled conditions [9]. The amorphous
regions are destroyed, leaving intact the crystalline segments.
Depending on the sources of the starting cellulose and on the

http://dx.doi.org/10.1016/j.ijbiomac.2017.03.076
0141-8130/© 2017 Elsevier B.V. All rights reserved.
554 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561

extraction process, various sizes of nanocrystals can be obtained.

CNC consist of needle-shaped nanoparticles with high crystallinity,
high surface area, average length of 100–200 nm and diameter in
the range of 5–10 nm [10]. The aspect ratio, defined as the ratio
between length and diameter, extends over a wide range [11]. Crys-
tallinity may vary from 65 to 95%, giving high strength, rigidity and
modulus, close to the theoretical modulus of perfect crystals [2].
The morphology and properties of CNC depend not only on the
source of the starting cellulose, but also on the extraction process
and on the characterization technique used [9]. Among the sources
used to obtain CNC we can quote cotton [12], curauá [13], byprod-
ucts such as rice husk [3] and maize straw [4], marine plants [14]
and still bacterial sources [15]. As main fields of applications of the
CNCs, we can mention reinforcing fillers in polymers [14,16,17],
packaging [18,19], biomedics [20,21] and water treatment [22,23].
The exhausted bark of Acacia mearnsii (black acacia) obtained
after tannin extraction was the raw material used in this work.
Black acacia is one of the most important cultivated trees in the
state of Rio Grande do Sul, Brazil. This kind of acacia is of wide
industrial use and it provides various products used as raw mate-
rials in industry. Plant extracts rich in tannin and phenols, as well
as flocculants used in water treatment processes, filtering phar-
maceutical and chemical products are obtained from its bark [24].
Nowadays, the bark residue remaining after tannin extraction leads Fig. 1. Scheme to obtain cellulose nanocrystals (CNC).
to an environmental problem being generally burned. According to
the Brazilian Institute of Geography and Statistics (IBGE) [25] the
washed with deionized water to neutral pH. The second bleaching
production of black acacia bark was around 70.000 ton in 2014 and
step was the treatment of 150 mg pulp with 5.0 mL of 80% acetic
this amount was totally produced in the state of Rio Grande do Sul,
acid (v/v) and 0.5 mL of 70% nitric acid (v/v), at 120 ◦ C, for 15 min.
where the climate is adequate for growing this plant. Thus, the use
The residual materials were washed with ethanol 95%, and deion-
of exhausted black acacia bark to produce CNC can be an interest-
ized water to neutral pH. They were finally dried at 60 ◦ C to constant
ing alternative to give value and to help solving an environmental
weight. Samples prepared using the solvent extraction step were
problem caused by the tannin industry.
named SE (samples with Solvent Extraction) and samples prepared
This study aimed to isolate and to characterize CNC from
without the step of solvent extraction were named WSE (samples
exhausted acacia bark following two different experimental pro-
Without Solvent Extraction).
cedures – with and without solvent extraction − to evaluate the
importance of this step on the morphology and thermal stability of
the materials. 2.2.2. Preparation of cellulose nanocrystals (CNC)
The SE-cellulose and WSE-cellulose were mixed with sulfu-
ric acid 64% (w/w) at a ratio of 1:8.75 (g/mL), at 45 ◦ C. Different
2. Experimental hydrolysis times were tested: 15, 30, 60, 90, 120, 150 and 180 min.
Reactions were stopped by pouring the mixtures into a large
2.1. Materials amount of cold water. The nanocrystals were further ultrasonicated
for 15 min in an ultrasonic bath Thornton (USC-1400 model) with
Exhausted acacia bark was supplied by SETA S.A. (Estância frequency of 40 kHz, followed by a prolonged dialysis (regenerated
Velha/RS, Brazil). Hexane (Fmaia, Brazil), ethanol (Fmaia, Brazil), cellulose membrane Fisher, cut-off 10,000–14,000 Da) against pure
sodium hydroxide (Labsynth, Brazil), hydrogen peroxide (CAQ deionized water to neutral pH. According to Dong [26], this proce-
Química, Brazil), nitric acid (Fmaia, Brazil), acetic acid (CAQ dure ensured that all ionic materials were removed except the H3 O+
Química, Brazil) and tetra-acetylethylenediamine (TAED, activa- counterions associated with the sulfate groups on the surface of the
tor) (Acros Organics, New Jersey, USA) were used as received. All nanocrystals. Suspensions of SE-CNC and WSE-CNC were stored in
solvents and reagents were of analytical grade. a freezer.
Fig. 1 shows the flowchart of the steps used to obtain SE-CNC
2.2. Procedures and WSE-CNC. At each step the samples were named according to
the procedure used.
2.2.1. Isolation of cellulose
Cellulose isolation procedures were based on Rosa et al. and 2.3. Characterization
Rehman et al. [3,4]. The dried acacia bark was milled under liquid
nitrogen and sieved (35 mesh). Part of the powder was dewaxed 2.3.1. Acacia bark and cellulose
using a sequence of hexane/ethanol/water in a Soxhlet apparatus Structural differences among the samples were revealed using
(6 h each step). Both the dewaxed and the non-dewaxed bark par- an ATR-FTIR Nicolet 6700 spectrometer through spectra obtained
ticles were delignificated at 121 ◦ C, in autoclave (Stermax 20EHD), with 64 scans and a resolution of 2 cm−1 .
using a 5% aqueous NaOH solution with a 1:30 bark to liquor ratio Scanning electron micrographs of samples corresponding to
®
(g/mL), for 30 min. Then, they were washed with deionized water each isolation step were obtained using a JEOL microscope JSM
to neutral pH. In the sequence, two bleaching processes were fol- 6060 operating at 20 kV. The test specimens were attached to an
lowed. In the first one, the bark was treated with an aqueous aluminum stub and sputtered with gold.
solution of H2 O2 (2%) and TAED (0.2%), at pH 12, for 12 h, at 48 ◦ C Wide-angle X-ray scattering experiments were performed using
(liquor to pulp ratio 25 mL:1 g). After this step, the materials were a Siemens D500 diffractometer. SE-cellulose and WSE-cellulose
 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561
were scanned under reflection mode, using an incident X-ray of
CuK␣ with wavelength of 1.54 Å, at a step width of 0.05 min−1 , from
2␪ = 0 to 40◦ . The Segal method [27] was used to calculate the crys-
tallinity of the samples. The Scherrer equation was used to calculate
the crystal thickness [28].
Thermogravimetric analyses were performed with a Q600 SDT
TA Instruments equipment under inert atmosphere at heating rate
10 ◦ C min−1 and heated to 600 ◦ C. The mass of the samples was
about 7 mg. Data shown are the average of 3 analyses.
2.3.2. Cellulose nanocrystals (CNC)
Considering the same hydrolysis time, cellulose nanocrystals
obtained from ten independent experiments were combined before
carrying out the characterization procedures.
Light scattering measurements were made by using a
Brookhaven Instruments Spectrometer with a He-Ne laser (Coher-
ent) operating at the wavelength of 633 nm. A Glan-Thomson prism
was used a polarizer which was calibrated with toluene [4,29].
The intensities were correlated with a correlator BI-9000AT. The
REPES software was used to obtain the relaxation time distribu-
tions [30]. Samples were diluted and centrifuged at 3000 rpm for
30 min before analyses. The results shown are representative of 3
analysis for each sample.
Morphological characterization of the CNC was taken using a
JEOL JEM 1200FxII transmission electron microscope operating at
80 kV. A drop of the diluted suspensions was deposited on a carbon-
coated grid. The samples were stained with uranyl acetate solution.
Around 200 representative CNC particles were used to determine
their dimensions which were calculated with the support of the
Image Tools software.
The thermogravimetric analyses of the CNC followed the same
conditions as described for cellulose.
3. Results and discussion
3.1. Acacia bark and cellulose
As already reported, 26% of extractives were found in the
exhausted acacia bark even after the industrial hot water-based
process of tannin extraction [16]. The results showed that from the
total extractives, 16.8% were soluble in organic solvents and 9.5%
were soluble in water.
ATR-FTIR was used to evaluate the chemical changes in acacia
bark after the various chemical treatments. Fig. 2 shows the FTIR
spectra of some materials obtained after each treatment step. All
samples showed two main absorption regions, one in the range
3500–2700 cm−1 related to stretching of the OH and CH groups
and the other in the range 1800–800 cm−1 related to stretching
vibrations of other functional groups or to bending vibrations. The
bands observed in the range 1200–900 cm−1 are due to C O
stretching [31], the band at 1317 cm−1 is due to CH2 deformations
and the band at 1430 cm−1 is due to CH2 bending [4]. According
to Viera et al. [32], lignin has characteristic bands at 1512 and
1613 cm−1 , due to stretching vibrations of aromatic C C bonds. The
absence of these bands in SE and WSE-cellulose indicates that most
of the lignin was removed. As already reported [33], the ketone C O
stretching absorption of hemicellulose can be observed as a weak
band between 1765 and 1715 cm−1 . We can clearly find this band
in the spectrum of crude acacia bark. As the process advances to the
following steps, we observe that such band disappears, indicating
that hemicellulose is being removed. Considering that acetic acid
was used in the second step of bleaching, the appearance of a band
at 1716 cm−1 indicates that some acetylation had occurred in SE
and WSE-cellulose [32,34,35].
555
Fig. 2. ATR-FTIR spectra: (a) crude acacia bark, (b) SE-acacia, (c) SE-NaOH, (d) SE-B1,
(e) SE-cellulose and (f) WSE-cellulose. (See Fig. 1 for sample designation).
Before looking at the SEM images, it is important to look at
the images of the samples after each isolation step (Fig. 3). Macro-
scopic images of the samples illustrate that the chemical treatments
changed the appearance of the materials, which switched from a
brown to a pale-yellow color after the purification steps.
We can clearly see that only the first bleaching step (SE-B1) was
not enough for pulp purification because the material was still yel-
lowish due to the presence of some residual lignin. We also observe
that SE-cellulose showed finer particles than WSE cellulose. Even
after maceration, the WSE-cellulose had an agglomerated appear-
ance, probably due to the presence of waxes and other compounds.
Images of acacia bark after solvent extraction (SE-acacia) and after
pulping (SE-NaOH) were not included in Fig. 3 because they did not
present significant changes in visual appearance, still resembling
the ground acacia bark.
Through SEM images (Figs. 4 and 5), we can confirm that the
morphology of the samples was highly modified after the chemi-
cal processes to which they were subjected. In Fig. 4a we see that
the ground acacia bark particles show variable sizes and shapes.
Comparing Fig. 4b and c, we observe that before solvent extraction
the particle surfaces were covered by substances that were later
removed.
SEM images in Fig. 5 allow comparing the morphology of the
samples considering the use or not of solvents in the extraction
step. Samples obtained after the pulping step are shown in Fig. 5a
(SE-NaOH) and Fig. 5b (WSE-NaOH). The presence of adhered mate-
rial only on the surface of the sample not submitted to solvent
extraction (Fig. 5b) confirms the effect of the solvents in removing
soluble compounds. After bleaching 1 (Fig. 5c and d), it is possible to
observe that the surface holes are emptier in the sample submitted
to solvent treatment. Comparing the morphology of SE-cellulose
(Fig. 5e and f) with WSE-cellulose (Fig. 5g and h), it is evident that
the SE-cellulose particles showed smaller size. The larger size of
the WSE-cellulose particles can be probably due to a “glue effect”
of the extractable compounds. This corroborates with the visual
appearance of the WSE-cellulose shown in Fig. 3e, showing the
agglomeration of the fibers even after maceration.
X-ray diffraction is a method commonly used to evaluate the
degree of crystallinity of the materials. In the case of our sam-
ples, only cellulose is crystalline. Free hydroxyl groups present in
cellulose macromolecules tend to form intra- and intermolecular
hydrogen bonds, giving rise to various different crystalline arrange-
ments [36]. Fig. 6 shows the XRD pattern of WSE-cellulose (the
556 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561

Fig. 3. Pictures: (a) crude acacia bark, (b) ground acacia bark, (c) SE-B1, (d) SE-cellulose and (e) WSE-cellulose.

Fig. 4. SEM images: (a, b) acacia bark milled samples, and (c) after solvent extraction (SE-acacia).

pattern of SE-cellulose was already published [37]). We observed
the peak of greatest intensity around 2␪ = 22◦ which represents the
crystallographic plane (200) of cellulose. The absence of a doublet
in this peak indicates that the extracted pulp from acacia bark has
crystalline structure of type I cellulose [33,38]. The other important
peaks at 2␪ = 15◦ and 2␪ = 34.5◦ are assigned to the crystallographic
planes (101) and (040) of cellulose and the intensity and broaden-
ing of these peaks do not differ much from XRD patterns previously
reported for commercial microcrystalline cellulose and cellulose
[3,34,38] and are similar to the diffraction patterns found in the
literature for other wood species [39].
The degree of crystallinity was calculated using the Segal
method [27] and it was found 86% (±0.8) for WSE-cellulose. The
degree crystallinity of SE-cellulose had been previously determined
as 82% (±0.8) [37]. These values were higher than those found in the
literature considering cellulose sources such as wood (71%), potato
tuber (68%), rice straw (68%) [40], sugarcane bagasse (76%) [41], rice
husk (67%) [3] and maize straw (75,5%) [4]. The slight difference in
crystallinity between SE-cellulose and WSE-cellulose results from
the solvent extraction step.
The WSE-cellulose crystal size was found to be 48 Å as deter-
mined from the XRD curve by using the Scherrer equation. It was
verified that these crystals were slightly larger than those of SE-
cellulose (45 Å) [37].
Fig. 7 shows the thermal degradation behaviour of samples after
selected isolation steps.
We can observe in Fig. 7 that the degradation of acacia bark
occurs over a wide temperature range between 200 and 500 ◦ C,
with a multistep pattern, corresponding to the degradation of its
constituents. According to Kim et al. [42], hemicellulose degrades
between 180 and 350 ◦ C, lignin between 250 and 500 ◦ C and
cellulose between 275 and 350 ◦ C. For crude acacia bark, the tem-
perature of maximum degradation rate was found at 357 ◦ C, which
corresponds to cellulose degradation. The lower shoulder close to
300 ◦ C refers to hemicellulose, while the degradation of lignin is
hardly seen over the entire temperature range because it has a
low mass loss rate. A small peak near 500 ◦ C can be seen in the
degradation curve of crude acacia bark. Mészáros et al. also found
a small peak in this region in another variety of acacia [43]. How-
ever, no explanation was provided with respect to that degradation
step. We suppose that products formed during degradation at lower
temperatures could be responsible for the degradation at ∼500 ◦ C.
It was evident that, as the samples were being submitted to
the chemical treatments, the degradation temperature range was
narrowed until the pulp has become decomposed in a single
step, confirming that hemicellulose and lignin were completely
removed. Because of this, the temperature onset of degradation
increased with the treatment. However, it was verified that the
temperature of maximum degradation rate, which is referred to the
degradation of cellulose, decreased with the treatment, as shown
in Table 1. Probably, the presence of other components inhibited
the diffusion of species, thus retarding cellulose degradation.
 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561 557

Fig. 5. SEM images: (a) SE-NaOH, (b) WSE-NaOH, (c) SE-B1, (d) WSE-B1, (e, f) SE-cellulose in different magnifications and (g, h) WSE-cellulose in different magnifications.

Notably, the Tmax of SE-cellulose was the lowest one, even lower presence of substances bonded to its surfaces [44–46]. Then, the
than that observed for WSE-cellulose. Variations in the cellulose increase of the degradation temperature of WSE-cellulose may be
degradation temperature have been discussed in recent studies. explained by all these factors.
The changes may be influenced by crystallinity, size of crystals and
558 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561

Table 1
Thermogravimetric data of the samples.

Sample Tmax , ◦ C Residue at 600 ◦ C, %

Crude acacia bark 358 ± 1 26.9 ± 0.4

SE-acacia 361 ± 2 22.3 ± 0.4
SE-NaOH 361 ± 3 21.0 ± 0.5
WSE-NaOH 364 ± 3 22.5 ± 0.4
SE-B1 362 ± 2 16.7 ± 0.2
WSE-B1 365 ± 5 18.8 ± 0.2
SE-cellulose 343 ± 3 4.0 ± 0.4
WSE-cellulose 358 ± 5 9.0 ± 0.3

Fig. 8. Intensity of scattered light for SE-CNC samples at different hydrolysis times:
15, 30, 60, 90, 120, 150 and 180 min in geometries depolarized scattering – VH and
polarized scattering – VV.

3.2. Cellulose nanocrystals

Cellulose nanocrystals were obtained from SE-cellulose and

Fig. 6. XRD pattern of WSE-cellulose.
Fig. 7. DTG curves of products from different steps of cellulose isolation: crude
acacia bark, SE-acacia, SE-B1, SE-cellulose and WSE-cellulose.
The residual mass after heating at 600 ◦ C decreased progres-
sively with treatment. While crude acacia bark showed 27% residue
at the end of the analysis, pure SE-cellulose showed just 4% residue.
WSE-cellulose showed a slightly higher mass of residue (9%), which
can be explained by the presence of some compounds not extracted
by solvents at the beginning of the purification process.
In other words the solvent extraction step affected some prop-
erties of the acacia bark cellulose particles obtained such as size,
crystallinity and thermal degradation behaviour due to the pres-
ence of some residual compounds attached to these particles.
WSE-cellulose being named SE-CNC and WSE-CNC. Both kinds of
cellulose nanocrystals were isolated by acid hydrolysis giving aque-
ous CNC suspensions of about 15 mg mL−1 .
Light scattering was used to confirm the presence of CNC and
to evaluate the effect of hydrolysis time. CNC have the property of
depolarizing scattered light due to their anisotropy. By using static
light scattering one obtains the depolarization ratio (␳V ), defined as
the ratio between the depolarized scattered light intensity (IVH ) and
the intensity of polarized scattered light (IVV ). The depolarization
ratio obtained for the acacia bark SE-CNC was 0.026 which is con-
sistent with the values found for CNC isolated from cotton [47] and
from maize straw [4]. Therefore depolarization ratio is independent
of the source from which the CNCs are extracted as already stated
in the literature [4,47].
In Fig. 8 we observe the intensity of scattered light, at 90◦
angle, for the SE-CNC samples at different hydrolysis times. Higher
intensities happened at the times of 30 and 60 min of hydrolysis,
regardless of the geometry. From these results, 30 and 60 min were
selected as experimental times to be tested for the acid hydrolysis
of cellulose.
In dynamic light scattering, the distribution of relaxation times
A(␶) can be related to a distribution of molecular weights, particle
sizes, diffusion coefficients or other relevant physical quantities.
Fig. 9 shows the distributions of relaxation times of SE-CNC and
WSE-CNC in aqueous suspension obtained at 30 and 60 min of acid
hydrolysis. It is well known that the longer relaxation time of the
particle, the larger the hydrodynamic radius [48].
It was seen that WSE-CNC showed only one relaxation peak,
indicating that all nanostructures in the sample had their hydrody-
namic radius within the same size range. Increasing hydrolysis time
from 30 to 60 min decreased the hydrodynamic radius of the par-
ticles showing that the reaction conditions must be controlled to
obtain adequate products. So, the light scattering experiments indi-
cated that 30 min of hydrolysis was long enough for the isolation
of CNC from acacia bark cellulose under these conditions.
On the other hand, SE-CNC samples showed two peaks in the
relaxation time distribution. The peak of higher intensity corre-
sponded to the largest population of particles and presented longer
relaxation times, indicating larger values of hydrodynamic radius.
The other peak, much lower, was related to smaller structures
which were assigned to nanocrystals that had undergone cleavage
 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561
Fig. 9. Effect of hydrolysis time on the relaxation times distribution for SE-CNC and
WSE-CNC.
Table 2
Average size of cellulose nanocrystals obtained after 30 min of hydrolysis.
Sample Length (nm) Diameter (nm) Aspect Ratio
SE-CNC 149 ± 45 8±2 18.6 ± 5.6
WSE-CNC 104 ± 71 9±3 11.5 ± 7.8
during acid hydrolysis. For 60 min hydrolysis, the lower peak repre-
sented even lower hydrodynamic radius than for 30 min hydrolysis,
indicating that the cleavage reaction was still happening.
Considering the average relaxation time, no appreciable dif-
ferences were found among the samples through light scattering
analyses
Fig. 10 shows the TEM images of SE-CNC and WSE-CNC prepared
with 30 min of hydrolysis. It can be seen that the obtained struc-
tures were relatively isolated with defined contours, as expected.
No significant differences were observed in the morphology of
nanocrystals obtained with and without the extraction step with
solvents, indicating that this step is not important for this isolation.
The average sizes of the nanocrystals are shown in Table 2.
Within the statistical error, we find that there is no difference in
length and aspect ratio in samples obtained with and without the
extraction step. Aspect ratio values obtained for the acacia bark cel-
Fig. 10. TEM images for the CNC prepared with 30 min of hydrolysis: (a) SE-CNC (b) WSE-CNC.
559
Fig. 11. Thermograms obtained for SE-CNC and WSE-CNC.
lulose nanocrystals were similar to those obtained from rice husk
[5], maize straw [6], curauá [13] and kenaf bast fibers [49]. How-
ever, they were lower than those obtained from wood (around 25)
[50] and bacterial cellulose (around 94) [51].
CNC were also studied through thermogravimetry. DTG curves
of SE-CNC and WSE-CNC (Fig. 11) showed two peaks. The peak with
maximum at about 380 ◦ C refers to the expected decomposition
temperature of cellulose [52]. The peak at around 190 ◦ C is prob-
ably related to the cellulose chains that have suffered sulfonation
during acid hydrolysis. According to Teodoro et al. [53], replacing
the hydroxyl groups by sulfate groups in the acid hydrolysis step
decreases the activation energy for the degradation of cellulose,
making the sample less thermal resistant. The presence of these
groups promotes dehydration reactions and catalyzes the decom-
position of the cellulose. More sulfonated cellulose regions degrade
at lower temperatures, in this case around 190 ◦ C, while the regions
less accessible to the modification tend to be more thermally stable.
The ash content at 800 ◦ C was similar in both cases, with 24% for
SE-CNC and 25% for WSE-CNC.
Considering these results we can conclude that, despite affecting
the properties of cellulose particles obtained from acacia bark, the
solvent extraction step did not affect the properties of nanocrystals
obtained ultimately at the end from those cellulose particles. Under
560 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561
the action of the acid catalyst employed to hydrolysate amorphous
cellulose, the residual compounds attached to the chains of the WSE
cellulose were probably dissolved and removed.
4. Conclusions
Cellulose nanocrystals were obtained from the exhausted black
acacia bark in this work. Two methods were used to isolate pure
cellulose from this raw material: without a solvent extraction step
(WSE-cellulose) and using a solvent extraction step (SE-cellulose).
The removal of lignin and hemicellulose from the crude acacia was
accomplished using a totally chlorine free method as confirmed
by ATR-FTIR. The crystallinity indices determined for the obtained
celluloses (82% and 86%) were higher than those found in the liter-
ature for celluloses obtained from other sources and using other
processes. The WSE-cellulose showed larger and more agglom-
erated particles, higher residual mass after thermal degradation
and higher crystallinity compared to the SE-cellulose. These results
suggest the presence of substances adhered to the surface of the
WSE-cellulose particles.
Light scattering measurements indicated that 30 min of acid
hydrolysis was enough time for the isolation of cellulose nanocrys-
tals from acacia bark cellulose, showing aspect ratio in the range
10–20. No significant difference was observed in the morphology
of nanocrystals obtained with and without the extraction step with
solvents at the initial stage. The nanocrystals also showed similar
thermal stability.
Then, we conclude that the solvent extraction step is impor-
tant to obtain cellulose from exhausted acacia bark however this
step is irrelevant to obtain cellulose nanocrystals from this mate-
rial. Such result is very important since it represents lower costs
with solvents and a more environmental friendly process.
Acknowledgments
The authors thank FAPERGS for scholarship and grants, Federal
University of Rio Grande do Sul (UFRGS), Sul-Rio-Grandense Insti-
tute for Education, Science and Technology (IFSul − Sapucaia do
Sul) and SETA S/A.
References
[1] T. Rogalinski, T. Ingram, G. Brunner, Hydrolysis of lignocellulosic biomass in
water under elevated temperatures and pressures, J. Supercrit. Fluid 47
(2008) 54–63.
[2] H.-M. Ng, L.T. Sin, T.-T. Tee, S.-T. Bee, D. Hui, C.-Y. Low, A.R. Rahmat, Extraction
of cellulose nanocrystals from plant sources for application as reinforcing
agent in polymers, Compos. Part B 75 (2015) 176–200.
[3] S.M.L. Rosa, N. Rehman, M.I.G. Miranda, S.M.B. Nachtigall, C.I.D. Bica,
Chlorine-free extraction of cellulose from rice husk and whisker isolation,
Carbohyd. Polym. 87 (2012) 1131–1138.
[4] N. Rehman, M.I.G. Miranda, S.M.L. Rosa, D.M. Pimentel, S.M.B. Nachtigall, C.I.D.
Bica, Cellulose and nanocellulose from maize straw: an insight on the crystal
properties, J. Polym. Environ. 22 (2014) 252–259.
[5] S. Sheshmani, A. Ashori, F. Farhani, Effect of extractives on the performance
properties of wood flour-polypropylene composites, J. Appl. Polym. Sci. 123
(2012) 1563–1567.
[6] S.A.L. Morais, E.A. Nascimento, C. Melo, Análise da madeira de Pinus oocarpa
Parte I–Estudo dos constituintes macromoleculares e extrativos voláteis,
Revista Árvore 29 (2005) 461–470.
[7] L.B. Brasileiro, J.L. Colodette, D. Piló-Veloso, A utilização de perácidos na
deslignificação e no branqueamento de polpas celulósicas, Quím. Nova 24
(2001) 819–829.
[8] C. Baptista, D. Robert, A.P. Duarte, Relationship between lignin structure and
delignification degree in Pinus pinaster kraft pulps, Bioresour. Technol. 99
(2008) 2349–2356.
[9] D.F. Martins, A.B. Souza, M.A. Henrique, H.A. Silvério, W.P.F. Neto, D. Pasquini,
The influence of the cellulose hydrolysis process on the structure of cellulose
nanocrystals extracted from capim mombaço (Panicum maximum), Ind. Crops
Prod. 65 (2015) 496–505.
[10] B.S.L. Brito, E.V. Pereira, J.-L. Putaux, B. Jean, Preparation, morphology and
structure of cellulose nanocrystals from bamboo fibers, Cellulose 19 (2012)
1527–1536.
[11] Y. Habibi, L.A. Lucia, O.J. Rojas, Cellulose nanocrystals: chemistry,
self-assembly and applications, Chem. Rev. 110 (2010) 3479–3500.
[12] A.A. Oun, J.-W. Rhim, Effect of post-treatments and concentration of cotton
linter cellulose nanocrystals on the properties of agar-based nanocomposites
films, Carbohyd. Polym 134 (2015) 20–29.
[13] D.O. Castro, E. Frollini, A. Ruvolo-Filho, A. Dufresne, Green Polyethylene and
curauá cellulose nanocrystal based nanocomposites: effect of vegetable oils as
coupling agent and processing technique, J. Polym. Sci. Pol. Phys. 53 (2015)
1010–1019.
[14] F. Bettaieb, R. Khiari, A. Dufresne, M.F. Mhenni, M.N. Belgacem, Mechanical
and thermal properties of Posidonia Oceanica cellulose nanocrystal reinforced
polymer, Carbohyd. Polym. 123 (2015) 99–104.
[15] J. George, K.V. Ramana, A.S. Bawa, Siddaramaiah, Bacterial cellulose
nanocrystals exhibiting high thermal stability and their polymer
nanocomposites, Int. J. Biol. Macromol. 48 (2011) 50–57.
[16] T. Taflick, E.G. Maich, L.D. Ferreira, C.I.D. Bica, S.R.S. Rodrigues, S.M.B.
Nachtigall, Acacia bark residues as filler in polypropylene composites,
Polímeros 25 (2015) 289–295.
[17] M.R. Kamal, V. Khoshkava, Effect of cellulose nanocrystals (CNC) on
rheological and mechanical properties and crystallization behavior of
PLA/CNC nanocomposites, Carbohyd. Polym. 123 (2015) 105–114.
[18] E. Mascheroni, R. Rampazzo, M.A. Ortenzi, G. Piva, S. Bonetti, L. Piergiovanni,
Comparison of cellulose nanocrystals obtained by sulfuric acid hydrolysis and
ammonium persulfate, to be used as coating on flexible food-packaging
materials, Cellulose 23 (2016) 779–793.
[19] H. Yu, C. Yan, J. Yao, Fully biodegradable food packaging materials based on
functionalized cellulose
nanocrystals/poly(3-hydroxybutyrate-co-3-hydroxyvalerate)
nanocomposites, RSC Adv. 4 (2014) 59792–59802.
[20] M. Jorfi, E.J. Foster, Recent advances in nanocellulose for biomedical
applications, J. Appl. Polym. Sci. 132 (2015) 41719.
[21] J.-C. Liu, D.J. Martin, R.J. Moon, J.P. Youngblood, Enhanced thermal stability of
biomedical thermoplastic polyurethane with the addition of cellulose
nanocrystals, J. Appl. Polym. Sci. 132 (2015) 41970.
[22] A.W. Carpenter, C.-F. Lannoy, M.R. Wiesner, Cellulose nanomaterials in water
treatment technologies, Environ. Sci. Technol. 49 (2015) 5277–5287.
[23] X. Yu, S. Tong, M. Ge, L. Wu, J. Zuo, C. Cao, W. Song, Adsorption of heavy metal
ions from aqueous solution by carboxylated cellulose nanocrystals, J. Environ.
Sci. 25 (2013) 933–943.
[24] L.S. Charão, Polinização em Acacia mearnsii De Wild, Revista de Ciências
Agro-Ambientais 3 (2005) 92–109.
[25] Produção da Extração Vegetal e da Silvicultura (ISSN 0103-8435), vol. 29,
Ministério do Planejamento, Orçamento e Gestão. Instituto Brasileiro de
Geografia e Estatística−IBGE, Rio de Janeiro, Brasil, 2014.
[26] X.M. Dong, J.-F. Revol, D.G. Gray, Effect of microcrystallite preparation
conditions on the formation of colloid crystals of cellulose, Cellulose 5 (1998)
19–32.
[27] L. Segal, J.J. Creely, A.E. Martin, C.M. Conrad, An empirical method for
estimating the degree of crystallinity of native cellulose using the X-ray
diffractometer, Text. Res. J. 29 (1959) 786–794.
[28] A. Abbott, A. Bismarck, Self-reinforced cellulose nanocomposites, Cellulose 17
(2010) 779–791.
[29] R.C.C. Leite, R.S. Moore, S.P.S. Porto, Use of a gas laser in studies of the
depolarization of the Rayleigh scattering from simple liquids, J. Chem. Phys.
40 (1964) 3741–3742.
[30] K. Schillen, W. Brown, R.M. Johnsen, Micellar sphere-to-road transition in an
aqueous triblock copolymer system–a dynamic light-scattering study of
translational and rotational diffusion, Macromolecules 27 (1994) 4825–4832.
[31] B. Xiao, X.F. Sun, R.C. Sun, Chemical structural, and thermal characterization
of alkali-soluble lignins and hemicelluloses, and cellulose from maize stems,
rye straw, and rice straw, Polym. Degrad. Stab. 74 (2001) 307–319.
[32] R.G.P. Viera, G.R. Filho, R.M.N. Assunção, C.S. Meireles, J.G. Vieira, G.S. Oliveira,
Synthesis and characterization of methylcellulose from sugar cane bagasse
cellulose, Carbohyd. Polym. 67 (2007) 182–189.
[33] J.I. Morán, V.A. Alvarez, V.P. Cyras, A. Vazquez, Extraction of cellulose and
preparation of nanocellulose from sisal fibers, Cellulose 15 (2008) 149–159.
[34] X. Sun, R. Sun, Y. Su, J. Sun, Comparative study of crude and purified cellulose
from wheat straw, J. Agric. Food Chem. 52 (2004) 839–847.
[35] R. Zuluaga, J. Putaux, A. Restrepo, I. Mondragon, P. Gañán, Cellulose
microfibrils from banana farming residues: isolation and characterization,
Cellulose 14 (2007) 585–592.
[36] M.-C. Popescu, C.-M. Popescu, G. Lisa, Y. Sakata, Evaluation of morphological
and chemical aspects of different wood species by spectroscopy and thermal
methods, J. Mol. Struct. 988 (2011) 65–72.
[37] M.I.G. Miranda, T. Taflick, S.M.L. Rosa, N. Rehman, A.T. Seefeld, V. Martins,
S.M.B. Nachtigall, C.I.D. Bica, Eco-friendly extraction of cellulose from
biorenewable sources, E-book (ISBN 978-987-24513-6-3), RIADICYP, Posadas,
Argentina, 2012.
[38] P. Bansal, M. Hall, M.J. Realff, J.H. Lee, A.S. Bommarius, Multivariate statistical
analysis of X-ray data from cellulose: a new method to determine degree of
crystallinity and predict hydrolysis rates, Bioresour.Technol. 101 (2010)
4461–4471.
[39] M. Poletto, A.J. Zattera, M.M.C. Forte, R.M.C. Santana, Thermal decomposition
of wood: influence of wood components and cellulose crystallite size,
Bioresour. Technol. 109 (2012) 148–153.
 T. Taflick et al. / International Journal of Biological Macromolecules 101 (2017) 553–561
[40] K. Abe, H. Yano, Comparison of the characteristics of cellulose microfibril
aggregates of wood, rice straw and potato tuber, Cellulose 16 (2009)
1017–1023.
[41] E.M. Teixeira, T.J. Bondancia, K.B.R. Teodoro, A.C. Corrêa, J.M. Marconcini,
L.H.C. Mattoso, Sugarcane bagasse whiskers: extraction and characterizations,
Ind. Crops Prod. 33 (2011) 63–66.
[42] H. Kim, S. Kim, H. Kim, H. Yang, Thermal properties of bio-flour-filled
polyolefin composites with different compatibilizing agent type and content,
Thermochim. Acta 451 (2006) 181–188.
[43] E. Mészáros, E. Jakab, C.J. Várhegyi, TG/MS, Py-GC/MS and THM-GC/MS study
of the composition and thermal behavior of extractive components of Robinia
pseudoacacia, J. Anal. Appl. Pyrolysis 79 (2007) 61–70.
[44] W. Chen, H. Yu, Y. Liu, P. Chen, M. Zhang, Y. Hai, Individualization of cellulose
nanofibers from wood using high-intensity ultrasonication combined with
chemical pretreatments, Carbohydr. Polym. 83 (2011) 1804–1811.
[45] F. Yao, Q. Wu, Y. Lei, W. Guo, Y. Xu, Thermal decomposition kinetics of natural
fibers: activation energy with dynamic thermogravimetric analysis, Polym.
Degrad. Stab. 93 (2008) 90–98.
[46] C. Vila, A.G. Barneto, A. Fillat, T. Vidal, J. Ariza, Use of thermogravimetric
analysis to monitor the effect of natural laccase mediators on flax pulp,
Bioresour. Technol. 102 (2011) 6554–6561.
[47] C.I.D. Bica, R. Borsali, C. Rochas, E. Geissler, Dynamics of cellulose whiskers
spatially trapped in agarose hydrogels, Macromolecules 39 (2006) 3622–3627.
561
[48] B.J. Berne, R. Pecora, Dynamic light scattering: with applications to chemistry,
biology and physics, 1st ed., 2000, Dover, New York.
[49] H. Kargarzadeh, I. Ahmad, I. Abdullah, A. Dufresne, S.Y. Zainudin, R.M.
Sheltami, Effects of hydrolysis conditions on the morphology, crystallinity,
and thermal stability of cellulose nanocrystals extracted from kenaf bast
fibers, Cellulose 19 (2012) 855–866.
[50] S. Beck-Candanedo, M. Roman, D.G. Gray, Effect of reaction conditions on the
properties and behavior of wood cellulose nanocrystals suspensions,
Biomacromolecules 6 (2005) 1048–1054.
[51] I.A. Sacui, R.C. Nieuwendaal, D.J. Burnett, S.J. Stranick, M. Jorfi, C. Weder, E.J.
Foster, R.T. Olsson, J.W. Gilman, Comparison of the properties of cellulose
nanocrystals and cellulose nanofibrils isolated from bacteria tunicate, and
wood processed using acid, enzymatic, mechanical, and oxidative methods,
Appl. Mater. Interfaces 6 (2014) 6127–6138.
[52] H. Yang, R. Yan, H. Chen, C. Zheng, D.H. Lee, D.T. Liang, In-depth investigation
of biomass pyrolysis based on three major components: hemicelluloses,
cellulose and lignin, Energ. Fuel 20 (2006) 388–393.
[53] K.B.R. Teodoro, E.M. Teixeira, A.C. Corrêa, A. Campos, J.M. Marconcini, L.H.C.
Mattoso, Whiskers de fibra de sisal obtidos sob diferentes condições de
hidrólise ácida: efeito do tempo e da temperatura, Polímeros 21 (2011)
280–285.