Preparation of cellulose nanomaterials via cellulose oxalates

Carbohydrate Polymers 213 (2019) 208–216

Contents lists available at ScienceDirect

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

Preparation of cellulose nanomaterials via cellulose oxalates
⁎

Jonatan Henschen , Dongfang Li, Monica Ek

⁎

T

Department of Fiber and Polymer Technology, KTH Royal Institute of Technology, Teknikringen 56-58, SE-100 44, Stockholm, Sweden

A R T I C LE I N FO

A B S T R A C T

Keywords:
Nanocellulose
Cellulose oxalate
Oxalic acid dihydrate
Cellulose
Cellulose nanofibrils
Cellulose nanocrystals

Nanocellulose prepared from cellulose oxalate has been discussed as an alternative to other methods to prepare

cellulose nanofibrils or crystals. The current work describes the use of a bulk reaction between pulp and oxalic
acid dihydrate to prepare cellulose oxalate followed by homogenization to produce nanocellulose. The prepared
nanocellulose is on average 350 nm long and 3–4 nm wide, with particles of size and shape similar to both
cellulose nanofibrils and cellulose nanocrystals. Films prepared from this nanocellulose have a maximum tensile
stress of 140–200 MPa, strain at break between 3% and 5%, and oxygen permeability in the range of 0.3–0.5 cm3
μm m−2 day−1 kPa−1 at 50% relative humidity. The presented results illustrate that cellulose oxalates may be a
low-cost method to prepare nanocellulose with properties reminiscent of those of both cellulose nanofibrils and
cellulose nanocrystals, which may open up new application areas for cellulose nanomaterials.

1. Introduction
As awareness of our impact on the planet is increasing, the demand
for new sustainable materials has grown rapidly in recent years, as
evidenced by the UN agenda 2030 (Assembly U.G., 2015). One raw
material that has the potential to be used in many new materials is
wood and cellulosic pulp fibre derived from wood. By isolation of the
nanostructures that make up wood fibres, it is possible to obtain materials with highly interesting properties, such as high specific strength
and stiffness (Tanpichai et al., 2012), high surface area (Moon, Martini,
Nairn, Simonsen, & Youngblood, 2011), biodegradability (Jung et al.,
2015) and low toxicity (Alexandrescu, Syverud, Gatti, & ChingaCarrasco, 2013).
The nanomaterials isolated from pulp fibres are generally categorized as either cellulose nanocrystals (CNCs) or cellulose nanofibrils
(CNFs). CNCs from wood are rod-like particles with a length of
50–500 nm and a width between 3 and 5 nm (Moon et al., 2011). These
particles are obtained through acid hydrolysis (Rånby, 1951) of cellulose fibres to remove all amorphous and less ordered regions in the
fibre, leaving only the crystalline regions of cellulose. This fabrication
process produces materials with high crystallinity ranging between 54
and 88% (Moon et al., 2011). CNFs, on the other hand, contain both the
amorphous and crystalline regions of the nanostructures in wood. This
makes the CNFs long and flexible with a length of 500–2000 nm and a
width between 4–20 nm (Moon et al., 2011). CNFs are produced
through mechanical refining of pulp fibres, which is usually preceded

by a chemical pre-treatment that reduces the adhesive forces between

⁎

fibrils and reduces the energy required to liberate them. One frequently
suggested application of CNFs and CNCs are as thin films. The difference in size between CNFs and CNCs makes the fabrication processes
different. CNFs can be prepared either by solvent casting or by filtration
over a fine membrane, while CNCs often is prepared by solvent casting.
Films prepared from these materials have high transparency (Aulin,
Salazar-Alvarez, & Lindström, 2012; Saito et al., 2009), high specific
strength (Saito et al., 2009; Syverud & Stenius, 2008) and are good
oxygen barriers. The properties of these films are however strongly
affected by humidity which for some cases means that they require
modification or mixing of the nanocellulose with other components to
reduce this effect (Aulin et al., 2012; Zhang, Zhang, Lu, & Deng, 2012).
Some suggested uses of these films are in food (Arora & Padua, 2010)
and electronic (Jung et al., 2015) applications.
In the pursuit of lowering production costs and changing the
properties of nanocellulose, many alternative pre-treatments and
methods to produce these materials have been published. CNFs have
been produced with varying pre-treatments that either reduce the cellulose chain length, e.g., enzymatic pre-treatment (Henriksson,
Henriksson, Berglund, & Lindstrom, 2007), or introduce electrostatic
charges to cellulose, e.g., TEMPO-mediated oxidation (Saito, Kimura,
Nishiyama, & Isogai, 2007), phosphorylation and carboxymethylation
(Wågberg et al., 2008). CNC has been produced with varying types and
concentrations of acid, including both organic and mineral acids, e.g.,
oxalic acid (Chen, Zhu, Baez, Kitin, & Elder, 2016; Li, Henschen, & Ek,
2017), citric acid (Spinella et al., 2016) and hydrochloric acid (Araki,
Wada, Kuga, & Okano, 1998). At low concentrations, these acids

Corresponding authors.
E-mail addresses: (J. Henschen), (D. Li), (M. Ek).

/>Received 15 November 2018; Received in revised form 20 January 2019; Accepted 16 February 2019
Available online 18 February 2019
0144-8617/ © 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
( />

Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

catalyse the hydrolysis of the amorphous regions of cellulose fibril aggregates. At higher concentrations, they can also react with cellulose to
introduce new functionality. Oxalic acid has been used both at low and
high concentrations to produce nanocellulose in different systems. The
first publication describing the production of nanocellulose using
highly concentrated oxalic acid used molten oxalic acid dihydrate (Ek
et al., 2015); later publications using oxalic acid as a 70% water solution (Chen et al., 2016) and the dissolution of oxalic acid in eutectic
solvents (Sirvio, Visanko, & Liimatainen, 2016) have appeared. Oxalic
acid dihydrate contains 29% water of crystallization and has a melting
point of 101 °C, making reactions with pulp fibres to produce cellulose
oxalates without additional solvent possible. The authors have previously published a short communication where molten oxalate dihydrate was used to prepare cellulose oxalate. This material was shortly
shown to be a suitable derivative to prepare CNCs at high yield using
ultrasonication (Li et al., 2017). The current paper further investigates
the possibility to prepare high-quality cellulosic nanomaterials using
oxalic acid dihydrate from different types of pulp, reaction conditions
and more sustainable washing procedures. Using microfluidization,
nanocellulose was prepared with particles resembling both CNFs and
CNCs. These materials characterized and found suitable to prepare films
using vacuum filtration with good mechanical and oxygen barrier

properties.

Table 1
The prepared samples, their names and reaction parameters. All reactions were
performed by mixing 20 g of pulp with 77 g of oxalic acid dihydrate at 110 °C.
Name

Raw material

Reaction time (min)

Washing

SWD35E
SWD35A
SWD60E
SWD60A
SWK35E
SWK35A
SWK60E
SWK60A

Dissolving
Dissolving
Dissolving
Dissolving
Kraft pulp
Kraft pulp
Kraft pulp
Kraft pulp

35
35
60
60
35
35
60
60

Ethanol
Acetone
Ethanol
Acetone
Ethanol
Acetone
Ethanol
Acetone

pulp
pulp
pulp
pulp

(2014). Initially, all samples were passed through two large chambers
of 400 μm and 200 μm connected in series. Following this, they were
passed multiples times through a 200 μm and a 100 μm interaction
chamber connected in series. The samples were passed 1, 3 or 5 times
through the small chambers, and are denoted as 1p, 3p or 5p, respectively. The pressure was set to 925 bar when passing the samples
through the large chambers and 1,600 bar when passing through the

large chambers.
2.2.2. Characterization
Fourier transform infrared spectroscopy (FTIR) spectra of the cellulose oxalates were measured with a PerkinElmer Spectrum 2000 FTIR (PerkinElmer, USA) equipped with a heat-controlled single reflection
attenuated total reflection (ATR) accessory (Golden Gate heat-controlled); 32 scans were recorded for each spectrum.
The nanocellulose content, represented by the amount of colloidal
particles, was calculated for SWD35A_5p and SWK35A_5p. This was
done by diluting the nanocellulose to 2 g L−1 and dispersing them using
a T18 UltraTurrax (IKA, Germany) at 16,000 RPM for 10 min before the
suspensions were centrifuged at 2,500 RCF for 1 h. The solid content of
the supernatant was determined and compared to the solid content
prior to centrifugation. The total nanocellulose yield was calculated by
multiplying the nanocellulose content with the gravimetric yield.
The free carboxyl content (FCC) of each cellulose oxalate was determined by conductometric titration as previously described (Habibi,
Chanzy, & Vignon, 2006; Li et al., 2017). A suspension was prepared by
mixing 100 mg of each cellulose oxalate with 100 ml of water and 10 ml
of a 0.01 M NaCl solution. The suspension was then stirred for 1 h and
titrated with 0.01 M NaOH. All titrations were carried out under constant nitrogen bubbling. The calculation of the content of free carboxyl
groups is based on the equation below:

2. Experimental
2.1. Materials
Softwood sulphite dissolving pulp sheets with a cellulose content of
≥96 wt % were supplied by Domsjö fabriker (Aditya Birla, Domsjö,
Sweden). Softwood kraft pulp sheets (Imperial Anchor) with a cellulose
content of 84 wt % were supplied by Holmen AB (Iggesund, Sweden).
Oxalic acid dihydrate (≥99%) was purchased from Sigma Aldrich
(Merck KGaA, Darmstadt, Germany). Ethanol (96%) and acetone
(≥99.5%) were supplied by VWR International AB (Stockholm,
Sweden). Commercially available food-grade olive oil was acquired at a
local supermarket (ICA, Stockholm, Sweden).

2.2. Methods
2.2.1. Preparation of cellulose oxalate
The dry pulp sheets were torn by hand into pieces of approximately
1 × 1 cm2. An evaporation flask was filled with 20 g of the torn pulp
and 77 g of oxalic acid dihydrate. To improve mixing, 30 glass balls (Ø:
15–16 mm) were added to the flask. The flask was attached to a rotary
evaporator and lowered into a heating bath set to 110 °C while rotated
for either 35 or 60 min. After the desired time was reached, the mixture
was lifted from the oil bath and allowed to cool at room temperature
while still rotating. After cooling down and solidifying, either ethanol
or acetone was added to dissolve excess oxalic acid, and the solution
was allowed to mix overnight. The added solvent was removed by vacuum filtration, and the remaining solids were washed further by filtering either acetone or ethanol through the material until the filtrate
reached a conductivity below 10 μS cm−1. At that point, the remaining
solids were collected as cellulose oxalate and allowed to dry at 40 °C in
an oven. The product was then obtained as a dry powder. The different
samples are denoted as seen in Table 1. The gravimetric yield of cellulose oxalate was calculated based on the dry weight of the pulp fibres
and cellulose oxalate.
To prepare nanocellulose out of cellulose oxalates, two of the
samples (SWK35E and SWD60E) were dispersed in deionized water at a
concentration of approximately 2 wt %. To fully dissociate the carboxylic acids and aid the fibrillation, the pH of the dispersions was
adjusted to pH 9–10 using 0.1 M sodium hydroxide before they were
mechanically disintegrated using a microfluidizer (M-110EH,
Microfluidics Corp, United States)similarly as described by Khan et al.

Free carboxyl content =

CNaOH × VNaOH
m

where CNaOH is the exact concentration (mol/l) of the NaOH solution,

VNaOH is the exact volume (l) of the NaOH solution used for titration
before the conductivity increased from the plateau of the titration curve
and m is the dry weight (g) of the oxalate sample.
The morphologies of SWK35E_3p and SWD60E_5p were studied by
adsorbing the samples on silica wafers and imaging them using scanning electron microscopy (SEM) and atomic force microscopy (AFM).
The samples were adsorbed onto the silicon wafers using a binding
layer of polyethylenimine. After washing and plasma treating the wafers they were dipped in polyethylenimine solution (1 g L−1), rinsed
with deionized water, dried in a flow of nitrogen, quickly dipped in
dilute suspensions of nanocellulose, rinsed with deionized waster and
dried in a flow of nitrogen. Just before imaging, the samples using an S4800 field emission SEM (Hitachi, Tokyo, Japan), they were coated
with a 5 nm platinum/palladium coating in a 208HR high-resolution
sputter coater (Cressington, Watford, UK) to suppress specimen charging. AFM images were acquired using tapping mode on a Multimode
IIIa instrument (Bruker, Santa Barbara, CA, United States). The images
209


Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

were recorded under ambient conditions (23 °C and 50% relative humidity) in air using Scanassyst cantilevers (Bruker, Santa Barbara, CA,
United States) with a nominal resonance frequency of 70 kHz and a
spring constant of 0.4 N m−1. The images captured by AFM were used
to determine the dimensions of the nanocellulose. The width was
measured using Nanoscope Analysis software (Bruker, Santa Barbara,
CA, United States) by measuring the height of 223 (SWK35E_3p) and
307 (SWD60E_5p) particles. The length was measured using ImageJ
(Schneider, Rasband, & Eliceiri, 2012) by tracking 89 (SWK35E_3p) and
59 (SWD60E_5p) particles.
Thermogravimetric analysis (TGA) was performed on a Mettler

Toledo TGA/SDTA 851e to study the thermal properties of the cellulose
oxalates. Approximately 4 mg of each sample was heated from 30 to
600 °C at a rate of 10 °C min−1 under N2 flow (flow rate 50 ml min−1).
The data were analysed by Mettler-STARe software.
X-ray diffraction (XRD) was performed on a PANalytical X’Pert PRO
MRD X-ray diffractometer (The Netherlands) equipped with a
PANalytical X’Celerator detector. Diffractograms were collected in a
range of 2θ = 10 to 50°. Cu Kα radiation was monochromatized with a
nickel filter. The crystallinity index was estimated as described by
Segal, Creely, Martin, and Conrad (1959).

Fig. 1. Potential esterification reaction of the cellulose with oxalic acid dihydrate; during the reaction, the cellulose also undergoes acid-catalysed hydrolysis (not shown).

more sterically available for chemical reactions, which makes them
more labile than those in cellulose in acid-catalysed hydrolysis and
dehydration. By washing the acid/pulp mixture with filtration it is
possible to eliminate the need for dialysis, this also facilitates recycling
the acid as it will be diluted less during washing. Both high demands for
chemical recovery and long washing procedures are two mayor obstacles when preparing conventional sulphuric acid based CNCs (Nelson
et al., 2016). Preparation of CNCs using hydrochloric acid can also be
done without the need for dialysis, unlike cellulose oxalate, the resulting material however has low surface charge and subsequently poor
colloidal stability (Yu et al., 2013).
Oxalic acid has the potential to participate in the simultaneous
hydrolysis and esterification of cellulose (Fig. 1) (Li et al., 2017), due to
the fibre structure it is difficult for the acid to fully penetrate the fibre
and react with all cellulose chains. Esterification is most likely to occur
on C6-OH due to its greater steric accessibility than C2-OH and C3-OH
(Fox, Li, Xu, & Edgar, 2011). Esterification with a difunctional carboxylic acid can result in both the introduction of charged acid groups
and crosslinking between cellulose chains. Nevertheless, our previous
work has indicated that most reacted oxalic acid leaves a charged group

on cellulose (Li et al., 2017).
The oxalate functionality contains two carbonyl groups, which can
be observed using FTIR. In all samples, a broad adsorption at
1739 cm−1 corresponding to the C]O stretching of the carbonyl group
was observed. This indicates successful esterification of the cellulose
oxalates. In Fig. 2, the spectra of the raw materials and SWK35E,
SWK35 A, SWD60A and SWK60A are shown; spectra for all samples are
found in the supplementary information (ESI).
The signal of carboxyl acids (−COOH) and ester bonds (−COO−)
overlaps and is difficult to differentiate in the FTIR spectra. Moreover,
no clear trends were observed in the intensities of the carbonyl signals
of the prepared samples. To determine the effect of the reaction time,
raw material, and washing liquor on the degree of esterification, conductometric titrations were performed to quantify the free carboxyl

2.2.3. Films
Films were prepared from nanocellulose suspensions (0.2 wt %)
dispersed using a T18 UltraTurrax (IKA, Germany) at 14,000 RPM for
20 min before they were degassed by sonication in an ULTRAsonik 28X
ultrasonic bath (NeyDental, Inc., Bloomfield, CT, USA) for 20 min.
Films of ˜50 g m−2 were produced using vacuum filtration over a
0.45 μm PVDF Durapore membrane filter (Merck Millipore) in a RapidKöthen sheet former (Paper Testing Instruments, Austria). Water was
drained from the suspensions until a stable gel-like material was left. At
this point, a second membrane was placed on top of the gel, and both
membranes were placed between two sheet-former carrier boards. The
whole assembly was dried for 20 min at 93 °C and 95 kPa reduced
pressure.
The oxygen permeability of the above CNF films was measured
using a Mocon Oxtran 2/20 (Modern Controls Inc., Minneapolis, MN)
instrument equipped with a coulometric sensor. The samples were
masked using aluminium foil, leaving a round area of 5 cm2 exposed for

the measurement. The measurements were conducted at 23 °C and at
50% RH. Each sample was conditioned prior to the measurement. Two
replicates for every sample were measured. The tensile properties of the
films were evaluated on an Instron Universal testing machine 5944
fitted with a 500 N load. The initial distance between the test grips was
20 mm, and the width of each specimen was 5 mm. At least 5 specimens
from each sample were tested. The separation rate was 2 mm min−1.
The samples were conditioned and tested at 23 °C and 50% RH. The
stress-strain curve was recorded for each test, and the data were averaged over all specimens. Young’s modulus was automatically calculated
by Bluehill version 3.72 (Illinois Tool Works Inc., Illinois, USA) software.
3. Results and discussion
Cellulose oxalates were prepared by mixing pulp with molten oxalic
acid dihydrate. As the reaction progressed, the appearance of the pulp
changed from a white to a greyish colour, followed by the torn pulp
sheet pieces losing their original shape as the mixture turned into a dark
yellow or brown paste. As the reaction cooled, the oxalic acid solidified,
and ethanol or acetone was added to remove excess oxalic acid. For the
experiments, both dissolving pulp and kraft pulp were used as raw
materials (Table 1). The observed colour change was more prominent
for samples prepared from kraft pulp, presumably due to its higher
amount of hemicellulose. Owing to the less ordered structural conformation of hemicellulose compared to that of cellulose (Cai, Zhang,
Charles, & Wyman, 2014), the hydroxyl groups of hemicellulose are

Fig. 2. FTIR spectra of a selection of the cellulose oxalates and raw materials.
The vertical line in the inset is at wavenumber 1739 cm−1 corresponding to the
C]O stretching of the carbonyl group. Spectra for all the remaining samples are
available in the supplementary information (ESI).
210

Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

was considerably lower at 0.62 mmol g−1, corresponding to an estimated degree of substitution (DS) of 0.13-0.16. No clear trends among
the FCCs of the cellulose oxalates washed with the same type of solvent
could be observed despite the differences in reaction time and type of
raw materials. As has been reported for the esterification of cellulose by
oxalic acid dihydrate, the DS on cellulose does not change when the
reaction time is prolonged from 30 to 60 min (Li et al., 2017). In our
previous work, we discussed that when no procedures for cellulose
activation (e.g., swelling) are carried out, substitution only occurs on
the accessible surfaces of cellulose fibrils. The highest theoretical DS of
cellulose in this case is 0.05 (Larsson, Hult, Wickholm, Pettersson, &
Iversen, 1999; Pu, Ziemer, & Ragauskas, 2006). In the current study, all
DS values exceeded this value, which indicates the occurrence of esterification on both accessible and inaccessible surfaces of the cellulose
fibrils. Nevertheless, as the reaction time increased from 35 min to
60 min, esterification could not proceed further. This could be attributed to the strong intermolecular interactions among the cellulose
chains, which limited the migration of molten oxalic acid further towards the crystalline regions of cellulose fibrils. Other methods to
prepare nanocellulose by reacting oxalic acid with pulp fibres produced
materials with FCC between 0.11 – 0.39 mmol g-1 (Chen et al., 2016;
Sirvio et al., 2016), this is considerably lower than the values observed
using the current procedure. It is believed that the higher FCC is obtained due to not adding water to quench the reaction. At elevated
temperatures, introducing additional water in this reaction is favorable
for the reverse reaction (hydrolysis on the formed ester bonds) but not
the forward reaction (esterification), due to the change in chemical
equilibrium. An increased FCC will facilitate the fibrillation when
preparing the nanocellulose.
In addition, the FCCs of samples washed with acetone were slightly
higher than those of samples washed with ethanol. During washing, it is

hypothesized that smaller fractions of cellulose oxalates with high DS
can interact rather well with polar solvents and thus can produce
smaller particles. The particles may be fine enough to either be entrapped in the filter paper or possibly pass through it. Consequently,
these fractions were removed from the samples after washing. As
ethanol is more polar than acetone, this is more likely to occur when

Table 2
The yield, free carboxyl content, degree of substitution and thermal degradation
onset temperature for the cellulose oxalates.

SWD35E
SWD35A
SWD60E
SWD60A
SWK35E
SWK35A
SWK60E
SWK60A

Yield [%]

Free carboxyl
content [mmol
g−1]

Estimated degree
of substitution

Thermal
degradation onset

temperature [°C]

93.8
97.1
85.4
96.4
83.6
96.0
85.6
99.3

0.86
1.05
0.62
1.08
0.92
1.10
0.97
1.04

0.15
0.18
0.11
0.19
0.16
0.19
0.17
0.18

175

176
175
175
177
173
176
N/A

Fig. 3. TG (left axis) and DTG (right axis) curves for SWD60A at 10 °C min−1
under N2 flow. The curves for the other samples are similar and are available in
the supplementary information (ESI).

content of each sample.
As shown in Table 2, the FCC of all samples ranged between
0.86–1.1 mmol g−1, except for SWD60E, which for unknown reasons

Fig. 4. Nanocellulose suspension at approximately 1.5 wt % prepared from the cellulose oxalates passed through the small chambers in the microfluidizer for either 1
passes (top) or 3 passes (bottom). SWK35E_3p was never collected.
211


Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

Fig. 5. Representative SEM (top) and AFM (middle) micrographs of nanocellulose adsorbed onto silica surfaces from suspensions of SWK35E_3p (left) and
SWD60E_5p (right). Distribution (bottom) width (left) and length (right) of SWD60E_5p and SWK35E_3p measured using AFM.

with ethanol resulted in lower yields in all samples, which could be
attributed to the removal of small fractions of hydrolysed and esterified

cellulose and hemicellulose during washing, as discussed above.
Thermal degradation of cellulose oxalates was detected by TGA to
determine the onset temperature and the maximum degradation temperatures of the main mass-loss regions. Three different mass-loss regions could be observed (Fig. 3). The first mass-loss region (30–130 °C)
could be attributed to the loss of physically adsorbed water in cellulose
oxalates (Moriana, Vilaplana, Karlsson, & Ribes, 2014). The thermal
degradation onset temperatures of all the studied cellulose oxalates
were 173–177 °C, which showed the beginning of the second main
mass-loss region (175–280 °C). This could be related to the decomposition of chemically attached oxalate groups as well as the dehydration of cellulose. The third main mass-loss region (280–375 °C) indicated the depolymerization of cellulose (Peng et al., 2013).
Considering the onset temperatures, the cellulose oxalates and

washing cellulose oxalate with ethanol.
As reported earlier (Li et al., 2017), prolonging the reaction time
from 35 min to 60 min when using dissolving pulp slightly decreases the
yield (Table 2). This is thought to be due to increased hydrolysis with
possible dissolution of some of the degradation products. This trend is
not observed when using kraft pulp as a raw material; for kraft pulp, the
yield increases slightly when increasing the reaction time from 30 min
to 60 min. There is no clear difference in the gravimetric yield between
the samples prepared from kraft or dissolving pulp, despite having
different cellulose contents. The dissolving pulp has a cellulose content > 96%, while kraft pulp contains 84% cellulose. This difference
indicates that the samples prepared from kraft pulp not only contain
cellulose oxalate but also hemicellulose, which has been hydrolysed
and esterified by oxalic acid.
The yield of cellulose oxalates was affected by whether either
ethanol or acetone was used to remove the excess oxalic acid. Washing
212


Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

aggregates present in the gel. This was determined for SWD35A_5p and
SWK35A_5p. After dilution and centrifugation, the content was determined to be 93% and 98% for SWD35A_5p and SWK35A_5p, respectively. The lower content for SWD35A_5p is likely connected to the
presence of larger aggregates, which reduce the viscosity and scatter
light, as observed when visually comparing the gels after homogenization. The total nanocellulose yield (from raw material to dispersed nanocellulose) was determined to 90% and 94% for SWD35A_5p
and SWK35A_5p, respectively.
Preparing nanocellulose from dry cellulose oxalate reduces the need
for biocides as it is possible to fibrillate the nanocellulose on demand. It
may also be open up new processes to fibrillate nanocellulose, for example in combination with other processing steps when preparing
formulations. Other studies have shown great potential to increase the
yield when producing nanocellulose by first producing CNCs followed
by preparing CNFs from the solid residue remaining after the reaction
(Chen et al., 2016; Wang, Zhu, & Considine, 2013; Wang, Chen, Zhu, &
Yang, 2017). These studies have used centrifugation followed by dialysis for washing the material, and in some cases required a substantial
number of passes in a microfluidizer to prepare the CNFs. The current
paper is able to simplify the washing step by using only filtration, and
to prepare a nanocellulose gel with only one pass through the 200 μm
and 100 μm interaction chamber of the microfluidizer. Simplifying the
process and reducing the required energy to fibrillate is crucial in order
to decrease the production cost in an industrial process.
Samples of SWK35A_5p and SWD35A_5p were used to produce
films; both of these samples formed even and transparent films with a
thickness of 30 μm when they were vacuum filtered and pressed
(Fig. 6). SEM images of the films (Fig. 7) show a closely packed fibril
structure. The SWD35A_5p and SWK35A_5p films had crystallinity indexes of 73.6% and 78.5%, respectively (Table 3). The crystallinity
indexes of the raw materials were measured to be 70.9% and 58.1% for
the dissolving pulp and kraft pulp, respectively. The crystallinity of
both samples increased after the reaction as amorphous regions in the
cellulose were hydrolysed, which was especially evident in the sample

prepared from kraft pulp.
Films produced from sulphuric or hydrochloric acid CNCs are often
produced using solvent casting because the particles are small enough
to pass through most filter membranes. Solvent casting is a very slow
process that, together with poor film properties, limits the use of films
produced from CNCs. The nanocellulose prepared in the current work
was prepared using vacuum filtration over a 0.43 μm membrane
without any substantial material loss. Films produced from SWD35A_5p
had a lower tensile strength and lower elongation at break compared to
those produced from SWK35A_5p (Table 3). In the present work, the
suspension of SWK35A_5p showed a higher transparency than the
suspension of SWD35A_5p at the same consistency, which indicates a
more efficient separation of nanosized cellulose fibrils in the former
than the latter (Moser, Lindström, & Henriksson, 2015). The more efficient separation of nanofibrils in the suspension enabled the exposure

Fig. 6. Films prepared from a) SWK35A_5p and b) SWD35A_5p produced by
vacuum filtration over a membrane.

presumably the resulting nanocellulose can be used as reinforcements
in thermoplastics, such as polystyrene (PS, melting temperature
74–105 °C), low-density polyethylene (LDPE, melting temperature
103–110 °C) and high-density polyethylene (HDPE melting temperature
125–132 °C).
When homogenizing the cellulose oxalates, they easily pass through
the 100 μm without clogging. At approximately 1.5 wt %, the samples
prepared from kraft pulp resulted in thick gels, while the samples
prepared from dissolving pulp were slightly less viscous and had a
higher opacity (Fig. 4). Gels could be made from all samples with high
viscosity after one pass through the microfluidizer. After homogenization of the cellulose oxalates, the difference in colour between the
different raw materials, as described above, is clearly apparent. Micrographs of the homogenized cellulose oxalates (Fig. 5) show that the

material consists of short fibrils with average lengths of 0.34 μm
(SWK35E) and 0.37 μm (SWD60E) and average widths of 3.2 nm
(SWK35E) and 4.3 nm (SWD60E). The width of the particles is comparable to that commonly found for CNCs and in the lower range for
CNFs. The length, however, has a large range that spans lengths typical
for both CNFs and CNCs. The analysed samples contain many particles
that are similar to the shape and length (50–500 nm) (Moon et al.,
2011) of CNCs and a considerable number of particles that are longer
(up to 1.1 μm) than common CNCs and shaped similar to flexible CNFs.
It should be noted that the number of long particles is likely underestimated because it is easier to adsorb, identify and measure small
particles than it is long and entangled particles when analysing the
images.
The nanocellulose content is an indication of the amount of larger

Fig. 7. SEM micrographs of films prepared from SWD35A_5p (left) and SWK35A_5p (right). Note the difference in magnification between the two images.
213


Carbohydrate Polymers 213 (2019) 208–216

0.14–5.03(Aulin, Gällstedt, & Lindström, 2010, 2012;
Naderi et al., 2016; Syverud & Stenius, 2008)

0.54
0.31
–

of more surface area for bonding between the nanofibrils during film
formation. SWK35 A also has a higher crystallinity index, which may
contribute to its higher mechanical strength. Consequently, the interfacial interactions between the nanofibrils were stronger in the film
made from SWK35 A than the one made from SWD35 A, as represented

by the superior tensile strength of the former relative to the latter. The
mechanical strength of the prepared films is comparable with many
other films prepared from CNFs, as is the strain at break and elastic
modulus.
In general, the films based on the nanocellulose in the current work
have comparable tensile properties to those reported for neat nanocellulose films (tensile strength: 30–240 MPa; elastic modulus:
1–17.5 GPa) (Alain, Jean-Yves, & Vignon, 1997; Fukuzumi, Saito,
Iwata, Kumamoto, & Isogai, 2009; Henriksson, Berglund, Isaksson,
Lindström, & Nishino, 2008; Iwamoto, Abe, & Yano, 2008; Iwamoto,
Nakagaito, & Yano, 2007; Iwamoto, Nakagaito, Yano, & Nogi, 2005;
Leitner, Hinterstoisser, Wastyn, Keckes, & Gindl, 2007; Henriksson &
Berglund, 2007; Nakagaito, Iwamoto, & Yano, 2005; Nakagaito & Yano,
2008; Nogi, Iwamoto, Nakagaito, & Yano, 2009; Rampinelli, Di Landro,
& Fujii, 2010; Saito et al., 2009; Stelte & Sanadi, 2009; Svagan, Samir, &
Berglund, 2007; Svagan, Hedenqvist, & Berglund, 2009; Syverud &
Stenius, 2008; Yano & Nakahara, 2004), as well as higher tensile
strength and elastic modulus than bio-based films based on non-cellulosic polysaccharides (e.g., starch, hemicelluloses, and pectin) (tensile
strength: 9.8–74 MPa; elastic modulus: 0.8–2.4 GPa) (Cao, Chen,
Chang, Stumborg, & Huneault, 2008; Edlund, Ryberg, & Albertsson,
2010; Le Normand, Moriana, & Ek, 2014; Mikkonen et al., 2010) and
commercial petroleum-based polymers (e.g., polyethylene, polypropylene, polyvinyl chloride, and polyamide) (tensile strength:
8–165 MPa; elastic modulus: 0.2–4.1 GPa) (Mangaraj, Goswami, &
Mahajan, 2009).
The films produced from SWD35A_5p and SWK35A_5p both showed
low oxygen permeability (Table 3), and the kraft-based film had slightly
lower oxygen permeability than the dissolving-based film. This can be
attributed to the higher degree of fibrillation and higher crystallinity
index of the samples produced from kraft pulp. Fewer large aggregates
increases the path required for gas molecules to travel through the film.
The reported permeability is comparable to that of many others at 50%

RH and is often regarded as sufficient for many applications.

2–10(Moon et al., 2011)

3.0 (+-0.4)
5.0 (+-0.6)
0.6(Reising et al., 2012)

4. Conclusions

66.2–82.1(Peng et al., 2013; Tejado, Alam, Antal,
Yang, & van de Ven, 2012; Zhao et al., 2013)

The high-yield preparation of cellulose oxalate through the reaction
between oxalic acid and cellulose was studied. During the reaction,
simultaneous esterification and acid hydrolysis occurs, resulting in a
highly charged (0.6–1.1 mmol/g) cellulose derivative with a crystallinity index of approximately 75%. The cellulose oxalate was prepared
as a dry powder, after homogenization all samples produces nanocellulose gels after one pass through the 100 μm interaction chamber of
the microfluidizer. The particle length of the resulting nanocellulose
varies greatly and contains particles which resemble the size and shape
of both CNF and CNC. It was shown that it is possible to prepare films
with mechanical and barrier properties similar to many other CNFs
through filtration. Using the described method it is possible to prepare
nanocellulose without the need for extensive dialysis or centrifugation
or extensive homogenization, and with very high yield. The nanocellulose has great potential for uses where the size of traditional CNFs
may present problems, such as when the particle length results in a too
high viscosity, and where CNCs are too small, such as when preparing
films though filtration.

SWD35A

SWK35A
Typical CNC
films
Typical CNF
films

74
79
54–88(Moon et al., 2011)

142 (+-5)
197 (+-7)
70(Reising, Moon, &
Youngblood, 2012)
95–240(Moon et al., 2011)

10.6 (+-0.4)
10.2 (+-0.4)
6–14.9(Moon et al., 2011;
Reising et al., 2012)
6-15(Moon et al., 2011)

Oxygen permeability [cm3 μm m−2 day−1 kPa−1] 50 %
RH
Modulus [GPa]
Tensile strain at maximum
tensile stress [%]
Maximum tensile stress [MPa]
Crystallinity index

Table 3
Crystallinity index, mechanical properties and oxygen barrier properties of the films prepared from SWD35A_5p and SWK35A_5p. Data are presented as the mean values. The error corresponds to the confidence interval,
alpha = 0.05.

J. Henschen, et al.

Conflicts of interest
The authors are shareholders in FineCell Sweden AB, a company
working in commercializing nanocellulose.
214


Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

Acknowledgments

Jung, Y. H., Chang, T., Zhang, H., Yao, C., Zheng, Q., Yang, V. W., ... Ma, Z. (2015). Highperformance green flexible electronics based on biodegradable cellulose nanofibril
paper. Nature Communications, 6, 7170. />Khan, A., Vu, K. D., Chauve, G., Bouchard, J., Riedl, B., & Lacroix, M. (2014).
Optimization of microfluidization for the homogeneous distribution of cellulose nanocrystals (CNCs) in biopolymeric matrix. Cellulose, 21(5), 3457–3468. https://doi.
org/10.1007/s10570-014-0361-9.
Larsson, P. T., Hult, E.-L., Wickholm, K., Pettersson, E., & Iversen, T. (1999). CP/MAS
13C-NMR spectroscopy applied to structure and interaction studies on cellulose I.
Solid State Nuclear Magnetic Resonance, 15(1), 31–40. />S0926-2040(99)00044-2.
Le Normand, M., Moriana, R., & Ek, M. (2014). The bark biorefinery: A side-stream of the
forest industry converted into nanocomposites with high oxygen-barrier properties.
Cellulose, 21(6), 4583–4594. />Leitner, J., Hinterstoisser, B., Wastyn, M., Keckes, J., & Gindl, W. (2007). Sugar beet
cellulose nanofibril-reinforced composites. Cellulose, 14(5), 419–425. https://doi.
org/10.1007/s10570-007-9131-2.

Li, D., Henschen, J., & Ek, M. (2017). Esterification and hydrolysis of cellulose using
oxalic acid dihydrate in a solvent-free reaction suitable for preparation of surfacefunctionalised cellulose nanocrystals with high yield. Green Chemistry, 19(23),
5564–5567. />Mangaraj, S., Goswami, T. K., & Mahajan, P. V. (2009). Applications of plastic films for
modified atmosphere packaging of fruits and vegetables: A review. Food Engineering
Reviews, 1(2), 133. />Mikkonen, K. S., Mathew, A. P., Pirkkalainen, K., Serimaa, R., Xu, C., Willför, S., ...
Tenkanen, M. (2010). Glucomannan composite films with cellulose nanowhiskers.
Cellulose, 17(1), 69–81. />Moon, R. J., Martini, A., Nairn, J., Simonsen, J., & Youngblood, J. (2011). Cellulose
nanomaterials review: Structure, properties and nanocomposites. Chemical Society
Reviews, 40(7), 3941–3994. />Moriana, R., Vilaplana, F., Karlsson, S., & Ribes, A. (2014). Correlation of chemical,
structural and thermal properties of natural fibres for their sustainable exploitation.
Carbohydrate Polymers, 112, 422–431. />009.
Moser, C., Lindström, M. E., & Henriksson, G. (2015). Toward industrially feasible
methods for following the process of manufacturing cellulose nanofibers.
BioResources, 10.
Naderi, A., Lindström, T., Flodberg, G., Sundström, J., Junel, K., Runebjörk, A., ...
Erlandsson, J. (2016). Phosphorylated nanofibrillated cellulose: production and
propertyes. Nordic Pulp and Paper Research Journal, 31(1), 20. />3183/npprj-2016-31-01-p020-029.
Nakagaito, A. N., & Yano, H. (2008). Toughness enhancement of cellulose nanocomposites by alkali treatment of the reinforcing cellulose nanofibers. Cellulose, 15(2),
323–331. />Nakagaito, A. N., Iwamoto, S., & Yano, H. (2005). Bacterial cellulose: The ultimate nanoscalar cellulose morphology for the production of high-strength composites. Applied
Physics A, 80(1), 93–97. />Nelson, K., Retsina, T., Iakovlev, M., van Heiningen, A., Deng, Y., Shatkin, J. A., &
Mulyadi, A. (2016). American process: Production of Low cost nanocellulose for renewable, advanced materials applications. In L. D. Madsen, & E. B. Svedberg (Eds.).
materials research for manufacturing: An industrial perspective of turning materials into
New products (pp. 267–302). Cham: Springer International Publishing.
Nogi, M., Iwamoto, S., Nakagaito, A. N., & Yano, H. (2009). Optically transparent nanofiber paper. Advanced Materials, 21(16), 1595–1598. />adma.200803174.
Peng, Y., Gardner, D. J., Han, Y., Kiziltas, A., Cai, Z., & Tshabalala, M. A. (2013).
Influence of drying method on the material properties of nanocellulose I:
Thermostability and crystallinity. Cellulose, 20(5), 2379–2392. />1007/s10570-013-0019-z.
Pu, Y., Ziemer, C., & Ragauskas, A. J. (2006). CP/MAS 13C NMR analysis of cellulase
treated bleached softwood kraft pulp. Carbohydrate Research, 341(5), 591–597.
/>Rampinelli, G., Di Landro, L., & Fujii, T. (2010). Characterization of biomaterials based on

microfibrillated cellulose with different modifications. Journal of Reinforced Plastics
and Composites, 29(12), 1793–1803. />Rånby, B. G. (1951). Fibrous macromolecular systems. Cellulose and muscle. The colloidal properties of cellulose micelles. Discussions of the Faraday Society, 11, 158–164.
Reising, A. B., Moon, R. J., & Youngblood, J. P. (2012). Effect of particle alignment on
mechanical properties of neat cellulose nanocrystal films. J-for-Journal of Science &
Technology for Forest Products and Processes, 2(6), 32–41.
Saito, T., Hirota, M., Tamura, N., Kimura, S., Fukuzumi, H., Heux, L., ... Isogai, A. (2009).
Individualization of nano-sized plant cellulose fibrils by direct surface carboxylation
using TEMPO catalyst under neutral conditions. Biomacromolecules, 10(7),
1992–1996. />Saito, T., Kimura, S., Nishiyama, Y., & Isogai, A. (2007). Cellulose nanofibers prepared by
TEMPO-mediated oxidation of native cellulose. Biomacromolecules, 8(8), 2485–2491.
/>Schneider, C. A., Rasband, W. S., & Eliceiri, K. W. (2012). NIH Image to ImageJ: 25 years
of image analysis. Nature Methods, 9(7), 671–675.
Segal, L., Creely, J. J., Martin, A. E., Jr., & Conrad, C. M. (1959). An empirical method for
estimating the degree of crystallinity of native cellulose using the X-ray diffractometer. Textile Research Journal, 29(10), 786–794. />004051755902901003.
Sirvio, J. A., Visanko, M., & Liimatainen, H. (2016). Acidic deep eutectic solvents as
hydrolytic media for cellulose nanocrystal production. Biomacromolecules, 17(9),
3025–3032. />
This research was made possible through funding by Södra Research
Foundation, Sweden, Gunnar and Birgitta Nordin's foundation, Sweden
(grant no. GFS2016-0274) managed by The Royal Swedish Academy of
Agriculture and Forestry (KSLA) and Ångpanneföreningen's Foundation
for Research and Development (ÅFORSK), Sweden(grant no. 17-521).
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi: />References
Alain, D., Jean-Yves, C., & Vignon, M. R. (1997). Mechanical behavior of sheets prepared
from sugar beet cellulose microfibrils. Journal of Applied Polymer Science, 64(6),
1185–1194. />Alexandrescu, L., Syverud, K., Gatti, A., & Chinga-Carrasco, G. (2013). Cytotoxicity tests
of cellulose nanofibril-based structures. Cellulose, 20(4), 1765–1775. />10.1007/s10570-013-9948-9.
Araki, J., Wada, M., Kuga, S., & Okano, T. (1998). Flow properties of microcrystalline

cellulose suspension prepared by acid treatment of native cellulose. Colloids and
Surfaces A: Physicochemical and Engineering Aspects, 142(1), 75–82. />10.1016/S0927-7757(98)00404-X.
Arora, A., & Padua, G. W. (2010). Review: Nanocomposites in food packaging. Journal of
Food Science, 75(1), R43–R49. />Assembly U.G (2015). Transforming our world : the 2030 Agenda for Sustainable
Development. Retrieved from />transformingourworld.
Aulin, C., Gällstedt, M., & Lindström, T. (2010). Oxygen and oil barrier properties of
microfibrillated cellulose films and coatings. Cellulose, 17(3), 559–574. https://doi.
org/10.1007/s10570-009-9393-y.
Aulin, C., Salazar-Alvarez, G., & Lindström, T. (2012). High strength, flexible and
transparent nanofibrillated cellulose–nanoclay biohybrid films with tunable oxygen
and water vapor permeability. Nanoscale, 4(20), 6622–6628. />1039/C2NR31726E.
Cai, C. M., Zhang, T., Charles, R. K., & Wyman, E. (2014). Integrated furfural production
as a renewable fuel and chemical platform from lignocellulosic biomass. Journal of
Chemical Technology & Biotechnology, 89(1), 2–10. />4168.
Cao, X., Chen, Y., Chang, P. R., Stumborg, M., & Huneault, M. A. (2008). Green composites reinforced with hemp nanocrystals in plasticized starch. Journal of Applied
Polymer Science, 109(6), 3804–3810. />Chen, L., Zhu, J. Y., Baez, C., Kitin, P., & Elder, T. (2016). Highly thermal-stable and
functional cellulose nanocrystals and nanofibrils produced using fully recyclable
organic acids. Green Chemistry, 18(13), 3835–3843. />C6GC00687F.
Edlund, U., Ryberg, Y. Z., & Albertsson, A.-C. (2010). Barrier films from renewable forestry waste. Biomacromolecules, 11(9), 2532–2538. />bm100767g.
Ek, M., Henschen, J., & Li, D. (2015). SE Patent No. SE 539317.
Fox, S. C., Li, B., Xu, D., & Edgar, K. J. (2011). Regioselective esterification and etherification of cellulose: A review. Biomacromolecules, 12(6), 1956–1972. https://doi.
org/10.1021/bm200260d.
Fukuzumi, H., Saito, T., Iwata, T., Kumamoto, Y., & Isogai, A. (2009). Transparent and
high gas barrier films of cellulose nanofibers prepared by TEMPO-Mediated oxidation. Biomacromolecules, 10(1), 162–165. />Habibi, Y., Chanzy, H., & Vignon, M. R. (2006). TEMPO-mediated surface oxidation of
cellulose whiskers. Cellulose, 13(6), 679–687. />Henriksson, M., & Berglund, L. A. (2007). Structure and properties of cellulose nanocomposite films containing melamine formaldehyde. Journal of Applied Polymer
Science, 106(4), 2817–2824. />Henriksson, M., Berglund, L. A., Isaksson, P., Lindström, T., & Nishino, T. (2008).
Cellulose nanopaper structures of high toughness. Biomacromolecules, 9(6),
1579–1585. />Henriksson, M., Henriksson, G., Berglund, L. A., & Lindstrom, T. (2007). An environmentally friendly method for enzyme-assisted preparation of microfibrillated
cellulose (MFC) nanofibers. European Polymer Journal, 43(8), 3434–3441. https://doi.
org/10.1016/j.eurpolymj.2007.05.038.

Iwamoto, S., Abe, K., & Yano, H. (2008). The effect of hemicelluloses on wood pulp nanofibrillation and nanofiber network characteristics. Biomacromolecules, 9(3),
1022–1026. />Iwamoto, S., Nakagaito, A. N., & Yano, H. (2007). Nano-fibrillation of pulp fibers for the
processing of transparent nanocomposites. Applied Physics A, 89(2), 461–466. https://
doi.org/10.1007/s00339-007-4175-6.
Iwamoto, S., Nakagaito, A. N., Yano, H., & Nogi, M. (2005). Optically transparent composites reinforced with plant fiber-based nanofibers. Applied Physics A, 81(6),
1109–1112. />
215


Carbohydrate Polymers 213 (2019) 208–216

J. Henschen, et al.

Wågberg, L., Decher, G., Norgren, M., Lindström, T., Ankerfors, M., & Axnäs, K. (2008).
The build-up of polyelectrolyte multilayers of microfibrillated cellulose and cationic
polyelectrolytes. Langmuir, 24(3), 784–795. />Wang, Q. Q., Zhu, J. Y., & Considine, J. M. (2013). Strong and optically transparent films
prepared using cellulosic solid residue recovered from cellulose nanocrystals production waste stream. ACS Applied Materials & Interfaces, 5(7), 2527–2534.
Wang, R. B., Chen, L. H., Zhu, J. Y., & Yang, R. D. (2017). Tailored and integrated production of carboxylated cellulose nanocrystals (CNC) with nanofibrils (CNF) through
maleic acid hydrolysis. Chemnanomat, 3(5), 328–335.
Yano, H., & Nakahara, S. (2004). Bio-composites produced from plant microfiber bundles
with a nanometer unit web-like network. Journal of Materials Science, 39(5),
1635–1638. />Yu, H. Y., Qin, Z. Y., Liang, B. L., Liu, N., Zhou, Z., & Chen, L. (2013). Facile extraction of
thermally stable cellulose nanocrystals with a high yield of 93% through hydrochloric
acid hydrolysis under hydrothermal conditions. Journal of Materials Chemistry A,
1(12), 3938–3944. />Zhang, W., Zhang, Y., Lu, C., & Deng, Y. (2012). Aerogels from crosslinked cellulose
nano/micro-fibrils and their fast shape recovery property in water. Journal of
Materials Chemistry, 22(23), 11642–11650. />Zhao, J., Zhang, W., Zhang, X., Zhang, X., Lu, C., & Deng, Y. (2013). Extraction of cellulose nanofibrils from dry softwood pulp using high shear homogenization.
Carbohydrate Polymers, 97(2), 695–702. />050.

Spinella, S., Maiorana, A., Qian, Q., Dawson, N. J., Hepworth, V., McCallum, S. A., ...

Gross, R. A. (2016). Concurrent cellulose hydrolysis and esterification to prepare a
surface-modified cellulose nanocrystal decorated with carboxylic acid moieties. ACS
Sustainable Chemistry & Engineering, 4(3), 1538–1550. />acssuschemeng.5b01489.
Stelte, W., & Sanadi, A. R. (2009). Preparation and characterization of cellulose nanofibers from two commercial hardwood and softwood pulps. Industrial & Engineering
Chemistry Research, 48(24), 11211–11219. />Svagan, A. J., Samir, M. A. S. A., & Berglund, L. A. (2007). Biomimetic polysaccharide
nanocomposites of high cellulose content and high toughness. Biomacromolecules,
8(8), 2556–2563. />Svagan, A. J., Hedenqvist, M. S., & Berglund, L. (2009). Reduced water vapour sorption in
cellulose nanocomposites with starch matrix. Composites Science and Technology,
69(3), 500–506. />Syverud, K., & Stenius, P. (2008). Strength and barrier properties of MFC films. Cellulose,
16(1), 75. />Tanpichai, S., Quero, F., Nogi, M., Yano, H., Young, R. J., Lindström, T., & Eichhorn, S. J.
(2012). Effective young’s modulus of bacterial and microfibrillated cellulose fibrils in
fibrous networks. Biomacromolecules, 13(5), 1340–1349. />bm300042t.
Tejado, A., Alam, M. N., Antal, M., Yang, H., & van de Ven, T. G. M. (2012). Energy
requirements for the disintegration of cellulose fibers into cellulose nanofibers.
Cellulose, 19(3), 831–842. />
216