Carbohydrate Polymers xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Eco-friendly isolation of cellulose nanoplatelets through oxidation under

mild conditions
⁎
L. Chávez-Guerreroa, , S. Sepúlveda-Guzmána, J. Silva-Mendozab, C. Aguilar-Floresc,
O. Pérez-Camachod
a
Autonomous University of Nuevo León, Mechanical and Electrical Engineering School, Pedro de Alba s/n, San Nicolás de los Garza, Nuevo León, 66455, México
b
Autonomous University of Nuevo León, Chemistry School, Pedro de Alba s/n, San Nicolás de los Garza, Nuevo León, 66455, México
c
Papaloapan University, Chemistry School, Circuito Central #200, Parque Industrial, Tuxtepec, Oaxaca, 68400, México
d
Research Center for Applied Chemistry, Blvd. Enrique Reyna Hermosillo No. 140, Saltillo, Coahuila, 25294, México

A R T I C L E I N F O A B S T R A C T

Keywords: Agave is recognized as a low recalcitrant material, which makes it a potential source to obtain nanocellulose.
Nanocellulose Aqueous dispersions (in water, H2O2, H2O2/H2SO4) of agave powder were heated at 120 °C under vapor pressure
Agave (1 kg/cm2). The resultant materials were observed with an optical microscope (OM), a laser scanning microscope
One-pot (LSM) to obtain the thickness measurement and a scanning electron microscope (SEM) to observe morphology.
Cellulose nanofibers
Raman spectroscopy, X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) were used to obtain
Xerophyte
the chemical structure. Cellulose nanoplatelets (CNPs) from Agave salmiana were successfully isolated under
mild conditions. Physicochemical analysis indicates that lignin was removed in a single step oxidation with
hydrogen peroxide in presence of sulfuric acid at low concentration (0.17 M). The CNPs images revealed the
presence of entangled cellulose nanofibrils (Ø ≈ 14 nm) along the nanoplatelets (thickness ≈80 nm).

1. Introduction with studies on cellulose nanocrystals (CNC), bacterial cellulose (BC),

cellulose nanofibers (CNF) (Faradilla et al., 2017; Trache, Hussin,
Lignocellulosic biomass from land plants is the main cellulose Haafiz, & Thakur, 2017) and cellulose nanoplatelets (CNP) being pub-
source, the most abundant biopolymer, which comprises of a linear lished (Chávez-Guerrero et al., 2017). Several chemical approaches
chain of connected glucose molecules. The term cellulose was coined by (e.g., concentrated acid, ionic liquids, enzymatic degradation, and
the French chemist Anselme Payen in 1838 (Credou & Berthelot, 2014). TEMPO oxidation) and mechanical routes (e.g., high-pressure homo-
Since then, numerous in-depth studies have been conducted for easily genizer, ultrasound, and milling) have been developed to extract na-
obtaining this important material, producing minimal pollution in the nocellulose from plants; however most of these approaches comprise
process. Cellulose fibrils from plants comprise nanocrystals (3–20 nm in several steps, that require intensive use of energy, reagents in high
size) bound together through an amorphous phase, while lignin and concentrations, and large amounts of water and solvents (Isogai, Saito,
some proteins function as a glue that holds fibrils together, thereby & Fukuzumi, 2011; Rahimi, Ulbrich, Coon, & Stahl, 2014; Xiang & Lee,
producing microfibers or bundles that are responsible for the great 2000). Several studies have focused on improving the traditional
flexibility and endurance of higher plants (Mondal, 2017). methods of extracting cellulose (Chávez-Guerrero et al., 2017; Trache
Recently, several xerophyte plants from the agavoideae subfamily et al., 2017; Pérez-Pimienta et al., 2017) however, by changing the
have been studied as a source of cellulose for various applications source of the raw material (biomass), these processes can also be op-
(Pérez-Pimienta, López-Ortega, & Sanchez, 2017), mainly because of timized. Most of the approaches to obtain nanocellulose from agave use
their low lignin content [9.8%], compared with other cellulose sources, the same procedure as when wood it’s used, without taking into account
and low cell wall recalcitrance (Li et al., 2014). Traditionally, the agave that agave has a lower recalcitrance, where several pretreatments like
was used to produce fibers, food, and alcoholic beverages; however, bleaching or mechanical milling and high sulfuric acid concentration
recently it has been used to produce bioethanol (Li et al., 2014). Lately, [≈60%] during hydrolysis are applied to the raw material (Chávez-
research in the field of nanocellulose has been accelerating, particularly Guerrero et al., 2017; Pérez-Pimienta et al.,2017). Then, the

⁎
Corresponding author.
E-mail addresses: [email protected] (L. Chávez-Guerrero), [email protected] (S. Sepúlveda-Guzmán), [email protected] (J. Silva-Mendoza),
[email protected] (C. Aguilar-Flores), [email protected] (O. Pérez-Camacho).

https://doi.org/10.1016/j.carbpol.2017.11.100
Received 7 September 2017; Received in revised form 22 November 2017; Accepted 27 November 2017
0144-8617/ © 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: Chavez-Guerrero, L., Carbohydrate Polymers (2017), https://doi.org/10.1016/j.carbpol.2017.11.100
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

development of easier alternative methods to extract cellulose remains Table 1

a research goal of great importance, because nanocellulose its been Amount of reagents for each sample.
used in flexible devices (Hoeng, Denneulin, & Bras, 2016), transparent
CNPs Sample Matrix H2O H2O2 H2SO4
films (Faradilla et al., 2017), solar cells and liquid fuels (Somerville,
Youngs, Taylor, Davis, & Long, 2010) as well as a filler in polymers (g) (ml) (ml) (ml)
(Hoeng et al., 2016; Mondal, 2017). Agave shows considerable poten-
WA 1 200 0 0
tial as a source of cellulose owing to its worldwide abundance, with a
PA 1 197 3 0
reported efficiency of up to 34 MT/ha/year of biomass (Somerville SA 1 195 3 2
et al., 2010). This plant cultivated in semiarid lands is a low-cost
feedstock. It is characterized by low recalcitrance and is an important
source of glucose, galactose, sucrose, fructose and inulin (Li et al., standing transparent films by solution casting method (Fig. S7). Then,
2012). Hence, we aim to describe the extraction conditions to obtain the wet films were dried in an oven at 40 °C for 5 h.
purified nanocellulose from agave powder (parenchyma) in one-pot and
demonstrate the presence of platelet-like nanostructures comprising
nanocellulose (CNP) through a facile chemical procedure that involves 2.3. Chemical and physical characterization of cellulose nanoplatelets
the use of small amount of reagents and leads to the isolation of other
added-value chemicals (polysaccharides) besides cellulose. The chemical microstructure was further studied in a Thermo
Scientific K-Alpha X-ray Photoelectron Spectrometer System. The ana-
lysis was done with monochromatized Al Kα radiation
2. Materials and methods
(E = 1486.68 eV). Survey and C1s high-resolution spectra were re-
corded for all cellulose samples using an energy pass of 50 eV. Cellulose
2.1. Materials
films were fixed with double sided adhesive copper tape and placed in
the analysis chamber at 10−9 mBar of vacuum pressure. The charging
The sulfuric acid (95% w/w) and hydrogen peroxide (30% w/w),
effect was corrected by shifting the binding energies considering the
sodium hydroxide (pellets), 3,5 dinitrosalicylic acid, glucose powder,
C1s signal at 285 eV. Nonlinear fit, using Gaussian curves was per-
ethanol (anhydrous), deuterium oxide (99% D), were all of analytical
formed by maintaining the full-width at half-maximum (FWHM) con-
grade and purchased from Sigma-Aldrich. Deionized water was used for
stant for all components in a particularly high resolution spectrum.
all the experiments.
Raman spectra measurements were recorded at room temperature
using a Thermo Scientific DXR Raman microscope using a 532 nm laser
2.2. Cellulose extraction excitation. Samples were deposited on a glass microscope slide and
dried in an oven at 50 °C for 2 h. The spectral data were scanned for the
Lignocellulosic raw material from Agave salmiana leave was washed acquisition of up to 30 accumulations and 30 s laser exposure time.
with deionized water and freeze-dried for 48 h. Then, the lyophilized The morphology of samples was studied by SEM using a FEI NOVA
fragile agave samples (0.5 cm × 10 cm × 3 cm) were mechanically NANOSEM 200. The samples were observed with an accelerating vol-
milled (Bel Art 37250 Micro-Mill Grinder) for 30 s to separate the agave tage of 10 kV and a working distance of 5 mm.
into fiber (FA) and matrix (MA; Fig. S1). The MA was sieved (American A drop of the sample dispersion (WA, PA and SA) and a drop of
standard #150) and subjected to the oxidation reaction in order to ethanol were deposited on a silicon wafer, and then it was dried at 40 °C
remove non-cellulosic materials. The sieved MA samples were treated for 2 h. Then, the silicon wafer was glued with graphite tape on top of
with water (WA sample), hydrogen peroxide (PA sample) and low- an aluminum pin and coated with gold using a vacuum sputter coater.
concentration (0.17 M or 1.7 wt.%) sulfuric acid solution (SA sample), A laser scanning microscope (LSM) was used to obtain height pro-
as shown in Fig. 1 (Table 1). files with a ZEISS LSM 700, in order to study the topography of the
All dispersions were kept under stirring at 10 rpm for 10 min, then sample using a laser emitting at 405 nm. Samples were glued with an
were placed inside an autoclave and kept isothermally at 120 °C under adhesive carbon tape to a glass microscope slide. All images were ac-
1 kg/cm2 vapor pressure for 45 min. All the dispersions were separated quired at 100 x magnification at 512 × 512 pixels, using at least 150
by centrifugation and washed with deionized water until a pH 6 was layers to build each image.
reached. A WA, PA and SA suspension was used to obtain the free- A Leica light microscope DM 3000 model was used to determine the

Fig. 1. Schematic procedure of the isolated cellulose nano-

platelets (CNPs). WA, water-treated sample; PA, hydrogen-
treated sample; SA, dilute sulfuric acid solution-treated
sample.

2
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

CNPs shape. To enhance the contrast between the glass and lignin of cellulose extended from 1000 to 1200 cm−1, whereas the vibrational
present in the CNPs, a solution of 0.1% (w/v) of methylene blue was peaks of non-cellulosic carbohydrate polymers were observed at ap-
used. A drop of each sample (WA, PA and SA) was deposited on a glass proximately 1600, 1740 and 1660 cm−1, thereby making the region
microscope slide, and then it was dried at 40 °C for 2 h. Then, a drop of 1000–1200 cm−1 suitable for studying the effects of chemical treat-
the methylene blue solution was deposited on each sample, and then it ment on the cellulose structure (Gierlinger, Keplinger, & Harrington,
was rinsed with distilled water three times; finally it was dried at 40 °C 2012). The peak located at 1120 cm−1 was shared by non-cellulosic
for 2 h. carbohydrates. For this reason, it is useful in determining the lignin
The CNP’s were analyzed using a transmission electron microscope presence (Gierlinger et al., 2012; Jähn, Schröder, Füting, Schenzel, &
(TEM) FEI Titan G2 80–300 TEM/STEM. The sample SA was disperse in Diepenbrock, 2002). After the oxygen peroxide treatment, a peak at
water and then dried at 45 °C on a TEM grid, using a lacey carbon film 1147 cm−1 was also observed; thus, all the characteristic peaks of the
stained with 2% phosphotungstic acid in deionized water. The sample cellulose backbone were observed. The C]O stretching vibration peak
was observed at an accelerating voltage of 300 kV. at 1740 cm−1, which is present only in the water treated sample (WA)
The samples were analyzed using a Fourier transform infrared due to the carboxylate group (Table S1), was attributed to noncellulosic
spectrophotometer IRTracer-100 from Shimadzu equipped with atte- materials (besides lignin). The peak due to the stretching vibration of
nuated total reflectance (ATR) accessory. The extracted by-product the lignin functional group (C]C) was located at approximately
(Liquid by-product from SA) was dried and placed on a diamond crystal 1602 cm−1. Although such peak was still present after bleaching (in the
and the spectra were acquired in a wavelength range of hydrogen peroxide-treated sample, PA) it was totally eliminated after
4000–600 cm−1, with a spectral resolution of 4 cm−1. CNPs films were oxidation (as observed in the dilute sulfuric acid solution-treated
analyzed in transmission mode in wavelength range from 4000 to sample, SA), suggesting that oxidation proceeded with lignin extrac-
400 cm−1. The optical absorption spectrum was acquired with an tion. The spectrum of WA exhibited a broad peak around 1659 cm−1
Agilent Cary 5000 UV–vis-NIR. and an intense peak at 1463 cm−1, both of which were related to the
The samples were analyzed by X-ray diffraction using a D8 Advance vibrational modes of aromatic groups of lignin (Gierlinger et al., 2012).
Bruker powder diffractometer. The diffraction patterns were collected These peaks were only observed in WA. The low lignin content was
using a Cu Kα radiation (λ = 1.5406 Å) in a 2θ range from 10° to 40°, a remarkable, and it may explain the light color (Fig. S1) and the well-
step of 0.05° and a time of 0.7 s. The crystallinity index (CrI) was cal- established low recalcitrance of the raw material (Li et al., 2014), which
culated following the equation (Segal, Creely, Martin, & Conrad, 1959), results in facile cellulose purification.
CrI (%) = ((Itotal-Iam)/Itotal) ×100, where Itotal indicates the max- Furthermore, CNPs were analyzed via X-ray photoelectron spec-
imum intensity and Iam the minimum diffraction intensity (amorphous troscopy (XPS). Fig. S3 shows the survey scans for all cellulose samples:
contribution ≈18°). they exhibit the presence of the carbon and oxygen atoms associated
with cellulosic and non-cellulosic materials. The survey spectrum for
3. Results and discussion WA, suggests that the treatment of MA with hot water allows the ex-
traction of the water-soluble fraction (e.g., sugars) (Li et al., 2012).
3.1. Lyophilization process The survey spectrum of PA exhibited features that were attributed to
a similar composition compared to WA, but with a smaller contribution
The lyophilization process, to which the agave leaves were sub- of the peaks of Ca and N. However, a noticeable change was observed in
jected, played an important role in the successful separation of the fi- the survey spectrum of SA, a sample obtained after treatment with H2O2
bers from the polymeric matrix (MA; Fig. S1). In the current procedure, and diluted sulfuric acid. The spectrum of SA only displays the photo-
the ice crystals created inside the samples formed pores (Voda et al., emission peaks for C1s and O1s, which are mainly attributed to cellu-
2012). losic materials. For further analysis nonlinear fitting was performed for
This broke the chemical bonds, making it brittle, and facilitated the the high-resolution C1s spectra, the results are shown in Fig. 2b. The
delamination of the parenchyma cell walls (Fig. S1). The agave cell curve-fitting analysis of the XPS C1s spectrum of WA resulted in three
walls were ruptured after freeze drying, which results in water per- peaks. One peak at 285.0 eV is associated with the presence of non-
meating the cells freely during hydrolysis (Voda et al., 2012). After oxidized carbon atoms (CeH and CeC) arising from compounds, such
lyophilization, approximately 81 wt.% of the original agave sample was as lignin, proteins, and other extractive substances present in the cell
recovered in the form of ice (Fig. 1). Subsequently, water (originally wall of plants (Kafle, Shin, Lee, Park, & Kim, 2015). Two more photo-
inside the sample) was obtained as a byproduct, which offered the emission peaks at 286.7 and 288.3 eV were strongly associated with
opportunity for it to be used later in the process. Agave, as a xerophyte oxidized carbon units (CeO and C]O, respectively) in cellulosic ma-
plant growing in semiarid lands, can be used to store water when this terials (Cao, Wang, Zeng, & Shen, 2016). According to these results, the
specific method is used, obtaining an important resource in places difference in the C1s peak intensity between spectra of WA, PA, and SA
where water is a scarce commodity. Once the agave was dried (19 wt.% is due to differences in the composition of each sample. WA exhibits the
of the original material), it was mechanically separated into fibers (FA; highest contribution of the C1s peak at 285.0 eV (CeH and CeC), which
17.6 wt.%) and matrix (MA; 82.4 wt.%) which comprises many com- indicates a high concentration of lignin (Johansson, Campbell,
ponents, including parenchyma, stoma, non-structural carbohydrates, Koljonen, & Stenius, 1999). This conclusion is in agreement with the
and calcium oxalate (Fig. 1). presence of photoemission peaks due to N1 s in the survey scan (Fig.
In a preliminary study, we analyzed the water-soluble components S3), which can be attributed to the structural protein network pro-
extracted after the experiment with hot water (WA), and we found that tecting the inner network in the plant cell wall (Keegstra, 2010) because
the byproducts generated during the process to obtain nanocellulose the cellulose isolation process strongly depends on protein degradation
were not pollutants. Instead, they are useful compounds, comprising (Demarty, Morvan & Thellier, 1984). WA results from the sample
reducing sugars (Fig. S1), such as fructose and glucose (Fig. S2). treated with hot water; showed that only the most labile proteins and
reducing sugars can be extracted, without the significant removal of
3.2. Chemical structure lignin. On the other hand, oxidation promoted by H2O2 leads to the
degradation of structural proteins and the oxidation of lignin units,
Raman spectra of the WA, PA and SA samples are shown in Fig. 2, which in turn, results in a decrease in the contribution of nitrogen and
depicting the effect of the oxidation on the CNPs. In particular, other calcium associated with proteins and in a reduction in the intensity of
components were shown to be eliminated from the cellulose surface, the C1s peak at 285.0 eV (CeH and CeC) observed in the spectrum of
allowing an appropriated detection of the CNPs. The finger-print region PA. Furthermore, the oxidation by H2O2 in the dilute H2SO4 medium

3
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

Fig. 2. Microstructural characterization of the sam-

ples after chemical treatments by Raman spectra (a),
and nonlinear fitting high-resolution X-ray photo-
electron spectroscopy (XPS) C1s spectra of cellulose
samples (b). Reaction pathway for lignin oxidation
(c).

favors protein degradation, and, under these conditions, the oxidation lignin performed by the UV spectroscopy (Nair, Dhar, & Vinu, 2016).
and solubilization of lignin were promoted, which in turn, assisted the UV spectrum of the liquid by-products after the CNP purification shows
extraction of crystalline cellulose. This result demonstrates that oxida- an absorption band at 270 nm associated with aromatic compounds
tion of lignin promotes its depolymerization, which occurs due to a (Fig. 3b), which is also in agreement with the FTIR spectrum. Further
mechanism where the H2O2 in acid medium, converts lignin into low microstructural characterization of the samples was performed by X-
molecular weight oxidized compounds with aromatic units (Fig. 2c) Ray diffraction (XRD) and results indicated that the material corre-
(Rahimi et al., 2014; Xiang & Lee, 2000), through the CeC or CeO sponds to cellulose I (French, 2013), showing an increase in the crys-
cleavage (Fig. S4). This result is in agreement with the analysis of the tallinity index (CI) as lignin was removed. The CI was 43.4% in the case
C1s spectrum of SA, which exhibits a larger contribution of the oxidized of WA (Table S2), then a similar value of 43.9% for PA, suggesting that
carbon units (CeO and C]O), associated with cellulosic materials than the sole use of H2O2 will not remove lignin entirely, only with the use of
the C1s spectrum of PA (Coseri et al., 2015). sulfuric acid the lignin was properly removed, showing a CI of 58.2%
We performed additional characterization to the extracted by-pro- (SA). The XRD results are in agreement with the ones obtained by FTIR
ducts (Liquid by-product from SA) in order to prove that the removing (Fig. 3c), showing a strong peak around 1604 cm1 in the case of WA and
of lignin (Fig. 3a-b) occurs through lignin oxidation and subsequent PA, which is characteristic to the C]C bond, indicating the presence of
CeC/CeO cleavage. Fig. S4 (a) shows a possible mechanism based on lignin, while SA presents almost the same profile as WA and PA, with
the literature, resulting in aromatic carboxylic acid units in oxidized the absence of the aromatic compounds, suggesting the proper pur-
lignin. The FTIR spectrum of the SA sample extracts (Fig. S4 b), shows ification of the CNP’s.
the CeH aromatic stretching at 2933 cm−1 (Fig. 3a), the C]C A deep morphological study was performed for all CNPs samples
stretching in aromatic compounds at 1597 and 1497 cm−1, the CeO and it shows CNPs were covered by an amorphous layer of lignin (Fig.
stretching associated to primary alcohols at 1006 cm−1 and C]O S5a-c). When this layer was eliminated, a flat surface was obtained (Fig.
stretching in carboxylic and ether groups at 1730 cm−1. These results S5 g-i). Such a thin layer was used to protect cellulose from enzymatic
are in agreement with the work reported by Xiang and Lee (Xiang & attack (cellulase); using scanning electron microscope (SEM) images, it
Lee, 2000) about the oxidation of lignin using H2O2 in acidic media, was possible to associate the non-cellulosic materials deposited on the
and the CeC cleavage resulting in aromatic with carboxylic groups (Fig. nanoplatelets surface (WA). After oxidation with H2O2, some amor-
S4 a). In addition, there are several studies on extracted/degraded phous materials were removed and the microstructure was partially

4
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

Fig. 3. FTIR (a) and UV (b) results of the liquid byproducts

from SA. FTIR of all the samples indicating lignin extraction
(c).

revealed (Fig. S5 d-f). As discussed previously via Raman spectroscopy in Fig. 5(a–c), where WA was deposited at the bottom of the glass bottle
and XPS, the noncellulosic materials were removed after acid hydro- because of the extra weight associated with the non-cellulosic materials
lysis, indicating that the dense areas observed in Fig. S5 (a-c) were and the highly hydrophobic interaction of the lignin presented in the
related to those materials. In their absence, the appearance of the sur- sample.
face was smoother, revealing the nanofibrils. On the other hand, PA and SA, which mostly comprised cellulose,
showed a more hydrophilic character than WA. The intensity of the
color of the nanoplatelets in Fig. 5(a–c), qualitatively reflects the
3.3. Morphological characterization
amount of non-cellulosic components because the samples were stained
with methylene blue, which reacts with lignin. WA shows a purple color
An isolated CNP from SA was observed using a laser scanning mi-
than those of PA and SA, indicating a higher amount of non-cellulosic
croscope (LSM, Fig. 4a), to measure the thickness around the zone in-
components, as previously inferred through the analysis of Raman
dicated by the lines shown in Fig. 4(d). Height profiles of SA after the
spectroscopy and XPS. In addition, it was shown that using a simple
amorphous material was removed are shown in Fig. 4(e), indicating a
light microscope, it is possible to observe the planar nanostructure and
thickness of approximately 70–82 nm, which corroborated the presence
its evolution through the chemical process of lignin extraction. Also, it
of well-defined nanoplatelets (Fig. 4b), which exposed the randomly
is possible to obtain transparent films made of CNPs, using an en-
oriented nanofibers (Ø ≈ 14 nm) on the layer’s surface (Fig. 4c).
vironmentally improved route (Fig. 5d), which may have several ap-
Although the plant cell wall comprises of aligned bundles of crys-
plications in flexible devices, transparent films, and solar cells.
talline cellulose fibrils, CNPs comprise randomly entangled nanofibrils
Nanometric parenchyma cell walls are the tissues responsible for
(Fig. 4c), that form well-defined nanostructures (Fig. 4b). Unlike the
sugar and water storage in succulent plants. The polysaccharide bio-
lateral ordering of nanofibers present in most land plants, in this case,
synthesis in cell walls generates random structural features whose di-
the nanofibers were randomly entangled. In addition, CNPs with large
verse characteristics are influenced by both the evolutionary and en-
aspect ratios can be observed in Fig. S6 (58000 nm/72 nm ≈ 800); this
vironmental factors (Burton, Gidley, & Fincher, 2010). In this case,
parameter is important in several applications, such as in scaffolds,
parenchyma is not required to resist degradation; this could explain the
drug delivery, transparent thin films, and composites (Ming et al., 2017;
low amount of lignin in A. salmiana. This fact suggest that cellulose
Kang et al., 2017).
present in the parenchyma had a function different from that present in
Cellulose dispersions (0.9 mg/ml) of the samples after 24 h shown

5
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

Fig. 4. A LSM image of one nanoplatelet in a). A SEM image from the square indicated in (a) was obtained and represented in (b–d). In (e), height profiles and measurements of the lines 1
and 2 indicated in (d) are shown.

the microfibers, which mainly functions as structural support. In fact, time-consuming and required the use of an excessive amount of re-
this is applicable to most cellulose fibers in plants (e.g., cotton, hemp, agents, causing sample degradation, which resulted in only a partial
and silk) and trees (Kirby, Gunning, Waldron, Morris, & Ng, 1996). This characterization of the structure.
evidence can be explained on the basis CNPs are not exposed to the
environment because the leaves are covered by a thick layer of chlor-
ophyll and an external cuticle, avoiding any of contact of the CNPs with 4. Conclusions
microorganisms and fungi.
Fig. 6(a-b) depicts TEM images of an individual CNP, where struc- CNPs were obtained using parenchyma cell walls from agave leave,
tural details can be appreciated. The cellulose nanofibers (≈3 nm) can obtained in one-pot under mild conditions, using 1.7 wt.% of H2SO4
be seen along the layer, where the entangled nanofibers can be re- and 0.48 wt.% of H2O2 which is relatively low, compared to the usual
sponsible for the nanoplatelets stability, by keeping its shape even after acid concentration (64 wt.% of H2SO4) during hydrolysis. We demon-
the hydrolysis at 120 °C. Nanofibers from other lignocellulosic mate- strated that hydrogen peroxide mixed in dilute sulfuric acid favors
rials and several isolation processes were already reported, showing protein and calcium oxalate degradation, also, under these conditions;
similar dimensions to the ones reported here (Chávez-Guerrero et al., the oxidation and solubilization of lignin were achieved. These results
2017; Trache et al., 2017). The result suggest that a CNP is a nanos- show that the material resembles nanoplatelets (thickness ranging from
tructured material, composed of cellulose nanofibrils, but a CNP it is a 70 to 80 nm) and the non-cellulosic materials were effectively removed,
nanoparticle itself (70 nm thickness), according to the International revealing the entangled cellulose fibrils that are approximately ≈3 nm
Organization for Standardization [ISO], with one of the three dimen- in diameter.
sions in the nanoscale (1 nm to 100 nm). Our findings present a way to obtain sugars, such as fructose and
To the best of our knowledge, this is the first procedure that is re- glucose, as a residue from nanocellulose extraction, an approach that
ported to obtain cellulose nanoplatelets using an up-scalable method improves the efficiency of biomass utilization. In addition, the results
that involves a facile one-step hydrolysis procedure (Guerrero & may contribute to the understanding of the relation between cellulose
Guzmán, 2015; Guerrero & Guzmán, 2017) conducted under mild morphology and its properties, which may encourage research in new
conditions (1.7 wt.% of H2SO4 and 0.48 wt.% of H2O2). Similar planar applications for nanocellulose-based advanced materials.
structures using parenchyma cells from sugar beet pulp were previously
reported; however, the thicknesses of the structures were not measured.
Instead, the layered structure was destroyed in an attempt to isolate Author contributions
nanofibrils (Dinand et al., 1999). Another similar structure was isolated
from onions (Nakashima, Mizuno, Takabe, Fujita, & Saiki, 1997) J.S.M and C. A.F. carried out the nanocellulose extraction. O.P.C.
however; the thickness of this structure was not accurately obtained as performed NMR analysis. S.S.G contributes performing SEM analyses
since the goal was to extract the ultimate fibrils. Both approaches in- and XPS, also contributes to process the XPS data and to write the
volve pretreatment and several steps; they were labor intensive and manuscript. L.C.G. performed the light microscope, LSM, SEM and
Raman analysis, directed the study and wrote the manuscript.

6
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

Fig. 5. Sample dispersions at the same concentration (right)

and an individual CNP (stained) CNP (left) from (a) WA, (b)
PA, and (c) SA. Scale bar = 50 μm. Transparent films of a
suspension containing CNPs after each treatment (d).

Acknowledgment Appendix A. Supplementary data

To the CIIIA, CIIDIT, the UANL PAICyT IT465-15 and CONACyTfor Supplementary data associated with this article can be found, in the
the financial support to this work. Contributions of the reviewers for the online version, at https://doi.org/10.1016/j.carbpol.2017.11.100.
journal are gratefully acknowledged.

Fig. 6. An individual CNP from SA can be observed using

TEM, at low magnification (a) and high magnification in (b).

7
L. Chávez-Guerrero et al. Carbohydrate Polymers xxx (xxxx) xxx–xxx

References Kafle, K., Shin, H., Lee, C., Park, S., & Kim, S. (2015). Progressive structural changes of
Avicel, bleached softwood and bacterial cellulose during enzymatic hydrolysis.
Scientific Reports, 5, 1–15.
Burton, R., Gidley, M., & Fincher, G. (2010). Heterogeneity in the chemistry, structure Kang, X., Sun, P., Kuga, S., Wang, C., Zhao, Y., Wu, M., & Huang, Y. (2017). Thin cellulose
and function of plant cell walls. Nature Chemical Biology, 6, 724–732. nanofiber from corncob cellulose and its performance in transparent nanopaper. ACS
Cao, W., Wang, Z., Zeng, Q., & Shen, C. (2016). 13C NMR and XPS characterization of Sustainable Chemistry & Engineering, 5, 2529–2534.
anion adsorbent with quaternary ammonium groups prepared from rice straw, corn Keegstra, K. (2010). Plant cell walls. Plant Physiology, 154, 483–486.
stalk and sugarcane bagasse. Applied Surface Science, 389, 404–410. Kirby, A., Gunning, A., Waldron, K., Morris, V., & Ng, A. (1996). Visualization of plant cell
Chávez-Guerrero, L., Sepúlveda-Guzmán, S., Rodríguez-Liñan, C., Silva-Mendoza, J., walls by atomic force microscopy. Biophysical Journal, 70, 1138–1143.
García-Gómez, N., & Pérez-Camacho, O. (2017). Isolation and characterization of Li, H., Foston, M., Kumar, R., Samuel, R., Gao, X., Hu, F., et al. (2012). Chemical com-
cellulose nanoplatelets from the parenchyma cells of Agave salmiana. Cellulose, 24, position and characterization of cellulose for Agave as a fast-growing, drought-tol-
3741–3752. erant biofuels feedstock. RSC Advances, 2, 4951.
Coseri, S., Biliuta, G., Zemljič, L., Srndovic, J., Larsson, P., Strnad, S., et al. (2015). Li, H., Pattathil, S., Foston, M., Ding, S., Kumar, R., Gao, X., et al. (2014). Agave proves to
Correction: One-shot carboxylation of microcrystalline cellulose in the presence of be a low recalcitrant lignocellulosic feedstock for biofuels production on semi-arid
nitroxyl radicals and sodium periodate. RSC Advances, 5 [96927-96927]. lands. Biotechnology for Biofuels, 7, 50.
Credou, J., & Berthelot, T. (2014). Cellulose: From biocompatible to bioactive material. Ming, S., Chen, G., He, J., Kuang, Y., Liu, Y., Tao, R., et al. (2017). Highly transparent and
Journal of Material Chemistry B, 2, 4767–4788. self-extinguishing nanofibrillated cellulose-monolayer clay nanoplatelet hybrid films.
Demarty, M., Morvan, C., & Thellier, M. (1984). Calcium and the cell wall. Plant Cell & Langmuir, 33, 8455–8462.
Environment, 7, 441–448. Mondal, S. (2017). Preparation, properties and applications of nanocellulosic materials.
Faradilla, R., Lee, G., Arns, J., Roberts, J., Martens, P., Stenzel, M., & Arcot, J. (2017). Carbohydrate Polymers, 163, 301–316.
Characteristics of a free-standing film from banana pseudostem nanocellulose gen- Nair, V., Dhar, P., & Vinu, R. (2016). Production of phenolics via photocatalysis of ball
erated from TEMPO-mediated oxidation. Carbohydrate Polymers, 174, 1156–1163. milled lignin–TiO2 mixtures in aqueous suspension. RSC Advances, 6, 18204–18216.
French, A. (2013). Idealized powder diffraction patterns for cellulose polymorphs. Nakashima, J., Mizuno, T., Takabe, K., Fujita, M., & Saiki, H. (1997). Direct visualization
Cellulose, 21, 885–896. of lignifying secondary wall thickenings in zinnia elegans cells in culture. Plant & Cell
Gierlinger, N., Keplinger, T., & Harrington, M. (2012). Imaging of plant cell walls by Physiology, 38, 818–827.
confocal Raman microscopy. Nature Protocols, 7, 1694–1708. Pérez-Pimienta, J., López-Ortega, M., & Sanchez, A. (2017). Recent developments in
Chávez Guerrero L., Sepúlveda Guzmán S (2015). Proceso De separación de materiales Agave performance as a drought-tolerant biofuel feedstock: agronomics, character-
vegetales fibrosos en sus componentes de fibra y matriz. Patent application: MX/a/ ization, and biorefining. Biofuels Bioproducts and Biorefining, 11, 732–748.
2015/001927. Rahimi, A., Ulbrich, A., Coon, J., & Stahl, S. (2014). Formic-acid-induced depolymer-
Chávez Guerrero L. & Sepúlveda Guzmán S. (2017). Proceso para la producción de ization of oxidized lignin to aromatics. Nature, 515, 249–252.
nanopartículas de celulosa en forma laminar a partir de material vegetal fibroso. Segal, L., Creely, J., Martin, A., & Conrad, C. (1959). An empirical method for estimating
Patent application: MX/a/2017/004481. the degree of crystallinity of native cellulose using the X-Ray diffractometer. Textile
Hoeng, F., Denneulin, A., & Bras, J. (2016). Use of nanocellulose in printed electronics: a Research Journal, 29, 786–794.
review. Nanoscale, 8, 13131–13154. Somerville, C., Youngs, H., Taylor, C., Davis, S., & Long, S. (2010). Feedstocks for lig-
ISO International Organization for Standardization (2015). ISO technical specifications nocellulosic biofuels. Science, 329, 790–792.
ISO/TS 80004-2: nanotechnologies-Vocabulary- part 2: nano-objects. Trache, D., Hussin, M., Haafiz, M., & Thakur, V. (2017). Recent progress in cellulose
Isogai, A., Saito, T., & Fukuzumi, H. (2011). TEMPO-oxidized cellulose nanofibers. nanocrystals: Sources and production. Nanoscale, 9, 1763–1786.
Nanoscale, 3, 71–85. Voda, A., Homan, N., Witek, M., Duijster, A., van Dalen, G., van der Sman, R., et al.
Jähn, A., Schröder, M., Füting, M., Schenzel, K., & Diepenbrock, W. (2002). (2012). The impact of freeze-drying on microstructure and rehydration properties of
Characterization of alkali treated flax fibres by means of FT Raman spectroscopy and carrot. Food Research International, 49, 687–693.
environmental scanning electron microscopy. Spectrochimica Acta Part A: Molecular Xiang, Q., & Lee, Y. (2000). Oxidative cracking of precipitated hardwood lignin by hy-
and Biomolecular Spectroscopy, 58, 2271–2279. drogen peroxide. Applied Biochemistry and Biotechnology, 84(-86), 153–162.
Johansson, L., Campbell, J., Koljonen, K., & Stenius, P. (1999). Evaluation of surface
lignin on cellulose fibers with XPS. Applied Surface Science, 144–145, 92–95.