Selective oxidation of bacterial cellulose by NO2-HNO 3

Article abstract of DOI:10.1039/c3ra44516j

Bacterial cellulose (BC) needs to be selectively oxidized by different oxidants such as TEMPO-NaClO-NaBr and NaIO₄ in order to control its degradability in vivo. However, both of these two oxidants are difficult to remove and the oxidation process is often accompanied by undesirable degradation of the material. In this work, a highly selective and efficient method for producing oxidized bacterial cellulose (OBC) has been developed in the reaction with NO₂ in HNO₃ solution. Systematic characterization of the structure and property of the resulting OBCs demonstrated that the oxidation process not only avoided the secondary reaction but also significantly enhanced its mechanical properties, controllability of biodegradability and protein absorbability. Nuclear magnetic resonance (NMR) testing and Fourier transform infrared spectroscopy (FTIR) proved that the selective oxidation of BC took place only at C6 where primary alcohol groups were quickly oxidized to carboxyl groups, with a maximum oxidation degree achieved up to 32.54% after 40 h. The oxidation process not only retained the crystal structure of BC, but slightly increased its degree of crystallinity, thus contributing to the improvement of its strength and toughness. At the same time, OBC manifested more controllable degradability and higher protein adsorption than the original BC as a function of the oxidation time.

Full text of DOI:10.1039/c3ra44516j

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Selective oxidation of bacterial cellulose by NO –
2
HNO3
Cite this: RSC Adv., 2014, 4, 1630
a
a
a
b
a
Qiuyan Cui, Yudong Zheng,* Qinghua Lin, Wenhui Song, Kun Qiao
a
and Shumin Liu
Bacterial cellulose (BC) needs to be selectively oxidized by different oxidants such as TEMPO–NaClO–NaBr
and NaIO in order to control its degradability in vivo. However, both of these two oxidants are difficult to
4
remove and the oxidation process is often accompanied by undesirable degradation of the material. In this
work, a highly selective and efficient method for producing oxidized bacterial cellulose (OBC) has been
2 3
developed in the reaction with NO in HNO solution. Systematic characterization of the structure and
property of the resulting OBCs demonstrated that the oxidation process not only avoided the secondary
reaction but also significantly enhanced its mechanical properties, controllability of biodegradability and
protein absorbability. Nuclear magnetic resonance (NMR) testing and Fourier transform infrared
spectroscopy (FTIR) proved that the selective oxidation of BC took place only at C6 where primary
alcohol groups were quickly oxidized to carboxyl groups, with a maximum oxidation degree achieved up
to 32.54% after 40 h. The oxidation process not only retained the crystal structure of BC, but slightly
increased its degree of crystallinity, thus contributing to the improvement of its strength and toughness.
At the same time, OBC manifested more controllable degradability and higher protein adsorption than
the original BC as a function of the oxidation time.
Received 20th August 2013
Accepted 14th October 2013
DOI: 10.1039/c3ra44516j
www.rsc.org/advances
hydroxyl groups. Aer oxidation, BC presented better degrad-
ability in order to match the growth rate of new tissue without
1
Introduction
Bacterial cellulose (BC) has raised interest as a biomedical undermining its intrinsic properties for the application as a
17
material due to its high crystallinity, good water-holding tissue engineering scaffold. However, secondary reactions
1
–4
capacity as well as its biocompatibility. What makes it special such as degradation of the original material oen accompany
15,20
is its 3-dimensional (3D) network structure consisting of cellu- the oxidation process when using the oxidants above,
and
lose nanobres, which contribute to its excellent hydrophilicity this has impaired its mechanical properties as implanted scaf-
5–7
4
and mechanical properties. By taking advantage of its unique fold materials. In addition, oxidation of BC by NaIO is a rela-
1
6,18
properties, BC and its derivatives have been widely exploited as tively slow process compared with TEMPO–NaClO–NaBr.
8
biomaterials in various applications such as articial kidney,
The aim of the present study was to explore the oxidation of
BC by an alternative nitrogen dioxide (NO )–nitric acid (HNO
temporary skin substitute and cardiovascular tissue engi- solution, which, to the best of our knowledge, is a new method
neering, etc. However, the 3D nano-network of BC lacks that has never been applied to BC. It is hypothesized that a well-
micropores wide enough for bone cell cultivation. In order to controlled oxidation process with high selectivity can nely
overcome this drawback, Zaborowska and co-workers have tailor the degradability of BC and make it a more suitable
incorporated 300–500 mm paraffin wax microspheres into the biomaterial for use in degradable implants, wound dressing
fermentation process. Besides, BC has also been blended with and tissue engineered scaffolds. A systematic study on the
poly(vinyl alcohol) (PVA) so as to obtain better shape mainte- oxidation conditions and the effect upon the structure and
nance capacity. In order to improve its degradability in vivo, properties of oxidized BC was performed. The oxidation was
BC has been mostly selectively oxidized by TEMPO–NaClO– proven to be of efficient selectivity with few secondary reactions
9
1
10
articial vessel endoprosthesis, cartilage, wound healing,
2
3
)
11
12
13
14
15,16
17–19
NaBr,
or NaIO
4
through which the primary alcohol involved. Furthermore, the degradability in vitro of oxidized
groups of BC were selectively oxidized into carboxyl groups and bacterial cellulose (OBC) was enhanced without impairing its
mechanical properties. Finally, bovine serum albumin (BSA)
a
was employed as a model protein to evaluate the protein
School of Materials Science and Engineering, University of Science and Technology
Beijing, Beijing 100083, PR China. E-mail: zhengyudong@mater.ustb.edu.cn; Fax: adsorption behavior of BC and OBC since a protein layer formed
+
86-010-62332336; Tel: +86-010-62330802
from the serum in body uid or the wound exudates is necessary
b
Centre for Nanotechnology
& Regenerative Medicine, Division of Surgery &
Interventional Science, University College London, London, NW3 2QG, UK
1630 | RSC Adv., 2014, 4, 1630–1639
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
21
ꢀ
for adhesion of cells, such as broblasts, to the material when temperature and humidity (20 C, 65% RH) in order to evaluate
it is implanted in the body or used as a wound dressing. the oxidation impact on their mechanical properties. At least
ve specimens in the original hydrogel state were tested for
each group. The ultimate tensile strength (s ), elongation at

t
2
Methods and procedures
break (3
t
) and Young's modulus of elasticity (E
according to eqn (3)–(5) respectively:
t
) were obtained
2.1 Preparation of OBC
BC membranes supplied by Hainan Yida Food Co, Ltd were
p
s
t
¼
(3)
(4)
(5)
pretreated before oxidation: BC membranes were immersed in
bd
ꢀ
0
.1 M NaOH solution for 60 min at 90 C to remove the bacterial
L ꢁ L
0
cell debris, and then they were washed to a neutral pH with
de-ionized water. Before oxidation, BC membranes were placed
3
t
¼
ꢂ 100
L
0
in 40% nitric acid (HNO ) solution. OBC samples were obtained
3
s
¼ 3
E
t
by guiding nitrogen dioxide (NO
different periods of time as described below. NO
by reaction of copper powder with concentrated HNO
supply of NO was continuous during the oxidation process.
2
) into the HNO
3
solution for
was obtained
and the
2
where p is the maximum stress that the materials underwent, b
the width of the samples, d the thickness, L the original length
of samples, and L the nal length. Toughness (T) was obtained
according to integral calculation of the strain–stress curve.
3
0
2
OBC lms were repeatedly washed to a constant pH with
de-ionized water aer oxidation for 10, 20, 30, and 40 h.
2.5 Degradation in vitro
2.2 Characterization
OBC membranes were cut into discs with a diameter of 15 mm
The chemical structure of BC and OBC before and aer degra-
dation for different times was tested by Fourier transform
infrared spectroscopy (FTIR; NICOLET 750) with a range of
and degraded in phosphate buffered saline solution (PBS, pH ¼
ꢀ
7
.4 ꢃ 0.2) at 37 C for different times. The samples were taken
ꢁ
1
ꢁ1
out separately aer being immersed for x weeks (x ¼ 0, 1, 2, 3, 5,
frequency from 4000 to 450 cm and a resolution of 4 cm
.
7
, 9, 12, 15 and 20), washed with de-ionized water, and then
The selectivity of oxidation of BC at C6 was identied
1
3
freeze-dried and weighed – this was dened as the dry weight,
according to solid-state
(
C nuclear magnetic resonance
1
3
W . Each time, four identical pieces of material were simulta-
x
C NMR; AVANCE III 400) with a frequency of 5 kHz, and the
neously immersed into four centrifuge tubes lled with 15 ml
chemical shi ranged from ꢁ50 to 250 ppm.
PBS for statistical analysis. The weight loss ratio was measured
and calculated according to the method mentioned by Chen
X-ray diffractometry (XRD; D/MAX-RB) (20 kV, 40 mA) with
Cu Ka radiation (l ¼ 0.154 nm) was used to examine the crystal
structure of the samples. The range of diffraction angles (2q)
3
et al., and an equation for this is shown below:
ꢀ
was 10–40 . The crystallinity of BC before and aer oxidation
Weight loss ratio (%) ¼ [(W
0
ꢁ W
x
)/W
0
] ꢂ 100%
(6)
was calculated according to eqn (1) using the Jade 5.0 profes-
sional soware.
Crystallinity ¼ diffraction intensity of crystal/(diffraction intensity of
2.6 Thermogravimetric analysis
crystal + diffraction intensity of amorphous region) ꢂ 100%
(1)
Thermogravimetry (TG) of BC and OBC was tested to analyze
their respective thermal stabilities. TG curves of the dried
2.3 Oxidation degree of OBC
samples were recorded using
a TA Instruments model
NETZSCH STA 449F3. The samples were heated from 50 to
The oxidation degree (OD) of OBC was tested by neutralization
reaction with sodium hydroxide (NaOH) since the carboxyl
ꢀ
ꢀ
ꢁ1
7
00 C at a rate of 10 C min
in an air current of
3
ꢁ1
+
100 cm min . Derivative TG (DTG) curves express the weight-
groups of OBC at C6 can be ionized with H in water, which can
ꢁ
loss rate as a function of temperature.
be measured with OH . Potassium hydrogen phthalate
(
C H KO ) was used to adjust the concentration of NaOH which
8 5 4
was dened as C. The volume of NaOH solution consumed was 2.7 Protein absorbability
set as V; M and m are the relative molecular weight of OBC
Bovine serum albumin (BSA) was used to characterise the
monomer and the dry weight of OBC. The oxidation degree was
then calculated according to eqn (2):
absorbability of BC and OBC. BSA solutions with a dened
concentration were prepared using PBS with a pH of 7.4 ꢃ 0.2.
ꢁ
3
C ꢂ V ꢂ M
The pH values were adjusted by adding 0.1 mol dm (0.1 M)
OD ¼
ꢂ 100%
(2)
ꢁ3
m
NaOH or 0.1 mol dm (0.1 M) HCl solution. In all the BSA
ꢁ
3
solutions, the ionic strength is almost constant (about 4 ꢂ 10
ꢁ
1
mol dm ); therefore, the effect of ionic strength can be
2.4 Mechanical properties of BC before and aer oxidation
ignored. The BC and OBC membranes were cut into small
BC and OBC membranes (120 ꢂ 20 mm) were both subject to pieces (20 ꢂ 30 mm) and placed into test tubes containing BSA
ꢁ
1
tensile testing with a speed of 20 mm min
at room solutions with the same initial concentration. The test tubes
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 1630–1639 | 1631

View Article Online
RSC Advances
were gently shaken at 25 C. Aer absorption, the BSA concen- attributed to the formation of carboxyl carbons.
Paper
ꢀ
15,28
Meanwhile,
tration in the test tubes was measured using a UV-vis spec- the signal of the original C6 at 64 ppm is obviously weakened
trometer (HP 8453, USA) via the absorbance at 280 nm. The pH rather than completely removed aer oxidation. This indicates
in BSA solution was also recorded. The absorption amounts and that hydroxyl groups at C6 were only partially turned into
capacity were calculated by comparing the concentration before carboxyl groups aer selective oxidation for 20 h, which is
and aer absorption.
consistent with the result of the existence of C]O in the gluco-
pyranoside units analyzed by FTIR above (Fig. 1). Therefore, we
have proven that BC was involved in selective oxidation by
3
Results and analysis
2 3
NO –HNO solution and that a fraction of hydroxyl groups at C6
3.1 Changes of chemical structure
were changed into carboxyl ones during the oxidation process.
Fig. 1 compares the FTIR spectra of BC and OBC oxidized for
different times. The spectral curve of BC is almost similar to the
3
.2 Oxidation degree of OBC
3
17
ones reported by the groups of Chen and Li. For the original
The oxidation degree of cellulose is one of the key criteria to
evaluate the efficiency of oxidation and is usually measured
through FTIR quantitative analysis and conductometric titra-
tion.
membranes which are different from pulverous natural cellulose.
Therefore, these methods for determining the oxidation degree
are unsuitable for OBC. In this work, we used a method of
chemical neutralization to measure the oxidation degree since
OBC could be constantly ionized with H from its carboxyl groups
at C6 in aqueous solution, and H can be completely neutralized
by the OH of NaOH solution. OBC membranes were cut into
pieces and were put into de-ionized water with mechanical stir-
ring so that H could be sufficiently ionized and neutralized by
OH . The OBC water slurries were gradually converted to trans-
parent and viscous uid as the oxidation time of BC was
increased. A schematic diagram of the neutralization equation is
shown in Fig. 3 The oxidation degree of OBC calculated according
to eqn (1) increased as the oxidation time was extended, and such
an increase became faster aer the rst 20 h of oxidation reaction
ꢁ
1
BC, absorption peaks at 3370, 2900, 1426, and 1060 cm
correspond to the stretching vibrations of n(–OH), n(–CH ),
2
2
2,23
d (–CH ) and n(C–O) respectively.
Aer oxidation for 10 h,
s
2
20,27
However, in this experiment BC existed in the form of
ꢁ1
the characteristic band at 1720–1740 cm
groups due to the C]O stretching vibration appeared
Fig. 1b), which was much more efficient than oxidation by NO
gas in aphotic conditions. Aer oxidation for 20 h and more,
the intensity of the C]O stretching vibration increased and
became more and more distinct with increasing degree of
oxidation (Fig. 1c–e). This phenomenon is in accordance with
of carboxylate
24
(
2
25
+
+
ꢁ
26
that reported by Kim et al. It should also be noted that there is
no other obvious difference between the spectra of the BC and
OBC samples, conrming that such oxidation is a stable process
with high selectivity. All these characteristics described above
indicate the formation of carboxyl groups in the OBC.
+
ꢁ
Solid-state NMR spectra of BC and OBC aer oxidation for
20 h in Fig. 2 present further evidence of the selectivity of the
oxidation reaction occurring. As expected, signals of the 6 carbon
atoms in each glucopyranoside unit of BC existed respectively at
regular positions in the NMR spectra, similar to those reported in
(Fig. 4). This is because oxidation rst occurred at the surface of
the BC membranes, and the hydroxyl groups inside were hardly
oxidized, leading to a relatively lower speed of oxidation.
However, microbrils gradually aggregated into bundles result-
ing in larger pores within the network as the oxidation continued
further, similar to other oxidation methods reported previ-
27,28
the literature.
As shown in Fig. 2, a sharp signal of carbon C1
was at about 105 ppm, C2 at 73 ppm, C3 at 75 ppm, C4 at 89 ppm
and C5 at 82 ppm, whereas the signal of C6 with a primary
hydroxyl group appears at 64 ppm. Compared with the NMR
spectrum of BC, the signals of the carbons in the oxidized units of
OBC are hardly affected by the oxidation process, except for a new
board but weak signal that appears near 170 ppm, which is
17,25
ously.
2 3
This change enabled more NO dissolved in the HNO
solution to permeate into the membrane more easily than before,
contributing to a relatively faster speed of oxidation. When BC
was oxidized for 40 h, the oxidation degree grew up to 32.54%.
Oxidized cellulose containing 16–24% carboxylic groups is
commercially available in powder, gauze, and fabric forms,
®
among which Surgicel is an typical example that has been used
for surgical haemostasis in clinical studies.
29,30
Therefore, OBC
achieved through this method has a wider range of achievable
oxidation degree at a relatively shorter period of time and may
have potential to be used for a variety of biomedical applications,
such as haemostatic materials, if oxidation time is optimized.
3.3 Change of crystal structure
XRD patterns of BC before and aer oxidation for different time
are exhibited in Fig. 5a. For the original BC, three classical
ꢀ
ꢀ
ꢀ
peaks are found at diffraction angles of 14.6 , 16.5 and 22.5 .
They stand for the primary diffraction of the crystal planes of
(1–10), (110) and (200) respectively, which prove the structure of
Fig. 1 FTIR spectra of BC (a) and OBC oxidized for (b) 10 h, (c) 20 h, (d)
3
0 h, and (e) 40 h.
1632 | RSC Adv., 2014, 4, 1630–1639
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
13
Fig. 2 Solid-state C NMR spectra of BC and OBC.
Fig. 3 Ionization and neutralization of OBC.
Fig. 4 Oxidation degree of BC after oxidation for different times.
Fig. 5 (a) XRD patterns and (b) crystallinity of BC and OBC with
different degrees of oxidation.
31,32
the cellulose I crystal.
Aer oxidation for different times, the
diffraction peaks showed a slightly higher intensity compared
with BC, indicating that OBC still retained the crystal structure
of cellulose I with an unexpected small improvement of crys- which is thought to be an anomaly (Fig. 5b). This interesting
tallinity. Thus the oxidation process hardly impacted the crystal phenomenon may be explained by a removal or rearrangement
structure of BC.
The crystallinity of BC varied slightly in an unpredictable way and the formation of new hydrogen bonds. It is well known that
aer oxidation. For the batch of the original BC membrane NO is one of the oxidants for the oxidation of BC. However, BC
studied, its crystallinity was calculated to be about 75.74%. It exposed to NO in a sealed environment oxidizes slowly and
of the more vulnerable amorphous region of BC membranes
2
2
should be noted that the crystallinity of the BC membranes tends to degrade. When BC was subjected to HNO solution and
3
varied slightly from batch to batch. Aer oxidation, small oxidized by NO , the HNO functioned as a catalyst during the
2
3
increases of crystallinity of the OBC samples from 75.54 to oxidation reaction, which improved the oxidation efficiency. At
7.89% were foundd as oxidation time was extended to 40 h, the same time, degradation of samples was alleviated by HNO
except for a small decrease of 71.73% for BC oxidized for 20 h solution, leading to the small change in the crystallinity degree
7
3
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 1630–1639 | 1633

View Article Online
RSC Advances
Paper
Fig. 6 Effect of oxidation degree on (a) stress–strain curve, (b) ultimate tensile strength, (c) toughness, (d) elongation at break, and (e) ultimate
modulus of elasticity of the samples.
2 3
of the OBC. Thereby, oxidation by NO –HNO solution did not with an elongation of up to 18%. OBCs oxidized for different
deteriorate but improved the crystallinity of the BC slightly, times showed more or less similar stress–strain proles to that
resulting in a remarkable improvement of the mechanical of BC, but with longer elongation, higher fracture strength and
properties as follows.
stress-stiffening values at higher strains. The OBCs appeared to
deform more easily at the rst stage with lower stress and longer
strain, and then the stress stiffening occurred gradually at a
higher strain when the oxidation time was increased. Conse-
3.4 Mechanical property
Fig. 6a shows typical stress–strain responses of the BC and OBC quently, the 2nd stage of linear elastic behavior shied gradu-
membranes. All the materials underwent the 1st stage of linear ally at an even higher strain (about 25%) and ended at the
elastic behavior, stress stiffening, 2nd stage of linear elastic highest stress of 0.75 MPa and elongation of >40% at break for
behavior and nal collapse regimes during stretching, similar OBC with an oxidation degree of 32.54%. It is clear that the
to a typical response for a thermoplastic elastomeric material oxidation process could effectively soen and toughen the
and displayed hyperelastic characteristics. Aer the initial low- nanocellulose network. It should also be noted that the yielding
stress elastic deformation, remarkable increases of both stage of the samples gradually disappeared as the oxidation
strength and modulus of BC induced by large strain were seen time was prolonged.
aer elongation above 11%. Following on from such stress
The ultimate tensile strength, toughness and elongation at
stiffening, the 2nd stage of linear elastic behavior took place break of the BC and OBC samples in Fig. 6b–d shared similar
1634 | RSC Adv., 2014, 4, 1630–1639
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 7 (a) Mass loss of BC and OBC with different degrees of oxidation, (b) log–log plot of the mass loss versus the degradation time (t) for the
different cases of oxidation degree, and (c) exponent of degradation power law of BC and OBC.
1
10% increase) of OBC oxidized for at most 40 h (Fig. 6d). As a
trade-off, there is a small decrease of modulus at both 1st and
nd stages of elastic deformation with an increase of oxidation
2
degree. The initial low-stress elasticity depended on the brous
network structure in terms of the bre packing structure and
density of the interconnection nodes between the bres. The
oxidation could attack the vulnerable amorphous region of the
nanobres and the original interconnection nodes between the

bres within the network with evidence of a small increase of
crystallinity of the network measured previously in Section 3.3.
In the meantime, the generation of the high-strength bond C]
O at C6 during oxidation may also be attributed to the change of
mechanical properties of the OBC samples. The presence of
carboxylic acid groups enhanced the intermolecular hydrogen
bond between the cellulose chains. As a result, the initial
interconnected nodes became weak and more nanobres
packed together into a thick brous bundle network with large
Fig. 8 TG curves of BC and OBC.
25
pores, as reported in our previous work. This may explain the
growing trends as the oxidation time was prolonged. The reduction of the strength and modulus of OBC at the 1st stage of
ultimate tensile strength of BC was about 0.51 MPa which is at low-stress elasticity. The alignment and movement of the
13
17
the same level as reported by Zaborowska and Li, then it brous bundles and cellulose chains were involved in the stages
increased up to 0.75 MPa (about a 44% increase) aer oxidation of large-strain-induced stress-stiffening and the 2nd linear
(i.e. OBCs). In contrast, a decrease of the tensile strength of 2,3- deformation, contributing to the profound increase in the
dialdehyde bacterial cellulose (DABC) was reported due to ultimate strength and toughness. In summary, regardless of the
17
oxidation by periodate. The toughness increased substantially initial small reduction of strength and elastic modulus (Fig. 6e),
from 4.10 MPa of BC to 9.08 MPa (an approx. 121% increase) of aer oxidation by NO –HNO solution for different periods of
2
3
OBC with an oxidation degree of 32.54% (Fig. 6c). Elongation at time, the ultimate strength and toughness of OBC samples have
break also grew signicantly from 19% of BC to 40% (about a been substantially improved.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 1630–1639 | 1635

View Article Online
RSC Advances
Paper
Fig. 9 (a) Calibration curve of absorbance versus BSA concentrations, (b) absorbability of BC and OBC with different degrees of oxidation, and (c)
maximum absorption amount of BC and OBC samples.
3.5 Degradability of OBC
Aer oxidation, the mass loss of the OBC samples in PBS
followed an exponential increase with the degradation time
depending on the degree of oxidation, i.e. mass loss (t) f t .
Fig. 7a displays the mass loss kinetics of BC and OBC samples
oxidized for different times. As expected, the degradability of
OBC in PBS was signicantly improved by the oxidation
process. The mass loss of BC was much slower throughout the
degradation process in vitro because of the inherent dense
packing of stable cellulose chains in its crystal structure. Both
BC and OBC samples were subjected to swelling when
immersed in PBS. It has been proven that water molecules
weaken C–O–C bonds in molecular chains and that the C–O
n
The power law exponent, n, the slope of a linear tting between
2
log(mass loss) and log(t), and the tting coefficient R were
obtained and listed in Fig. 7b. Fig. 7c shows the proportionally
increasing correlation between the exponent, n, of the degra-
dation power law of BC and OBC and the degree of oxidation.
Obviously, the degradation rate was improved by the oxidation
process and BC oxidized for a longer time had the faster
degradability.
3
bonds started to break slowly, leading to a slow mass loss.
Comparing to the mass loss results of OBC obtained by
Provided that degradation began rst at the amorphous
2
5
2
5
NO
oxidation degree of about 30%, we found that OBC by NO
HNO solution degraded more slowly. This is because degra-
2
gas conducted by Peng et al. with an approximate
regions, more defects and broken bonds were generated
subsequently as more cellulose chains were exposed to PBS.
Carboxyl groups are easier to rotate around the bond of C5–C6
than hydroxyl groups since the carboxyl groups at C6 possess
higher energy and more reactivity, which made the bond
between the glucose unit ring structures of OBC more liable to
fracture than with BC. Consequently, more broken molecular
chains came off the base network materials which speeded up
the mass loss. As the degradation was carried on, the accel-
eration of degradation gradually slowed down. It is envisaged
that the degradation at this stage mainly occurred within the
crystal regions, which were more difficult to degrade in PBS
2
–
3
dation had already occurred simultaneously during the
oxidation process by NO2 gas, resulting in more structural
defects on the surface and inside of OBC membrane,
thus making it degrade more easily in PBS. However,
oxidation by NO
secondary reactions including degradation were alleviated
by HNO solution.
2
–HNO
3
solution was temperate and
3
3.6 Effect of oxidation on thermal stability
ꢀ
due to the denser and more ordered packing of the molecular The TG curves between 50 and 650 C of BC and OBC
chains so that fewer structural defects in the crystalline samples are depicted in Fig. 8. The degradation behavior of
structure remained.
OBC samples showed signicant differences from that of
1636 | RSC Adv., 2014, 4, 1630–1639
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 10 BSA absorbability mechanism of BC and OBC: (a) schematic of protein adsorption onto OBC surface; (b) and (c) zeta potentials of BC and
OBC; (d) contact angles of BC and OBC; and (e) pH of BC and OBC in de-ionized water.
the BC sample. The rst minor weight loss observed for 3.7 The effect of oxidation degree on protein absorbability
ꢀ
both BC and OBC samples, occurring from 50 to about 250 C,
Since a protein layer formed from the serum in body uid or
3
3,34
is attributed to membrane dehydration.
Physically adsor-
wound exudates is necessary for adhesion of cells such as
bed and hydrogen bond linked water molecules can be lost at
this stage. Increasing carboxyl group content leads to a
gradual increase in the amount of lost water. Similar changes
in the degradation behavior of BC were reported in litera-
21

broblasts to the material when it is implanted in the body or
used as a wound dressing, the protein adsorption property is
closely related to cell adhesion behavior. BSA was used as a
model protein to evaluate the protein adsorption behavior of
the membranes. The calibration curve (Fig. 9a) of absorbance
versus BSA concentrations under working conditions was
obtained by linear t. The effect of the degree of oxidation of
OBC on BSA adsorption onto the membranes was analysed with
BC as control, as presented in Fig. 9b. All materials underwent
adsorption, negative adsorption and then the adsorption
balance process and the BSA absorbability has been improved
overall by the oxidation process regardless of some random
variations. Fig. 9c shows the maximum adsorption of BC and
OBC during the adsorption process. Unexpectedly, OBC with a
3
4,35
tures.
The onset of thermal degradation (temperature at
ꢀ
5
% weight loss) was found to decrease from 287 to 243
C
for BC oxidized for 10 h. OBC with a degree of oxidation of
ꢀ
4–36
3
2.54% decreased to about 230 C. The second mass loss is
3
due to thermal degradation.
With its different degree
of oxidation, OBC shows a relatively lower thermal stability
and a more complex decomposition process due to the
presence of carboxyl groups. This phenomenon indicates
that the degradation of OBC samples requires less
energy, theoretically proving the better degradability of OBC
over BC.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 1630–1639 | 1637

View Article Online
RSC Advances
Paper
degree of oxidation of 11.57% absorbed the most BSA at the
stage of adsorption balance, indicating that such an amount of
carboxyl groups of OBC might t adsorption the best.
Acknowledgements
This study is nancially supported by National Natural Science
Hydrophilicity and electrostatic interaction are the main Foundation of China Project (Grant no. 51073024 and Grant no.
37
factors that affect protein adsorption. The schematic diagram 51273021), and the National Science and Technology Support
in Fig. 10a is proposed to explain the adsorption mechanism of Project of China (Grant no. 2011BAK15B04).
the OBC samples. The contact angles of the OBC membranes
decreased slightly aer oxidation, indicating that they were
slightly more hydrophilic than BC (Fig. 10d). This phenomenon
is consistent with the zeta potentials of BC and the OBC
References
samples shown in Fig. 10b and 10c. BC showed a negative
charge (ꢁ14.0 mV) which was increased aer oxidation. BC
oxidized for 20 h with an oxidation degree of 11.57% acquired
the lowest negative zeta potential (ꢁ36.2 mV). It is also well
1 A. Svensson, E. Nicklasson, T. Harrah, B. Panilaitis,
D. L. Kaplan, M. Brittberg and P. Gatenholm, Bacterial
cellulose as a potential scaffold for tissue engineering of
cartilage, Biomaterials, 2005, 26(4), 419–431.
2 N. Petersen and P. Gatenholm, Bacterial cellulose-based
materials and medical devices: current state and
perspectives, J. Microbiol. Biotechnol., 2011, 91, 1277–1286.
3 Y. Chen, T. Xi, Y. Zheng, L. Zhou and Y. Wan, In Vitro
Structural Changes of Nano-Bacterial Cellulose Immersed
in Phosphate Buffer Solution, J. Biomimetics, Biomater.,
Tissue Eng., 2011, 10(55), 55–66.
3
7
known that the isoelectric point of BSA is about pH 4.7, below
which BSA solution will take a positive charge. Aer oxidation
for some time, BC acquired carboxyl groups at C6 and would be
negatively charged in aqueous solution (Fig. 10a). When OBC
was immersed in BSA solution, the carboxyl groups of OBC
made the pH of the solution below the isoelectric point
(Fig. 10e). And then positively charged BSA was bound to be
absorbed onto the OBC membranes by electrostatic interaction
which was an apparently predominant driving force for protein
adsorption.
4 H. Backdahl, G. Helenius, A. Bodin, U. Nannmark,
B. R. Johansson, B. Risberg and P. Gatenholm, Mechanical
properties of bacterial cellulose and interactions with
smooth muscle cells, Biomaterials, 2006, 27, 2141–2149.
5
M. Iguchi, Review Bacterial cellulose—a masterpiece of
nature's arts, J. Mater. Sci., 2000, 35, 261–270.
4
Conclusions
In this work, a selective and efficient oxidation method for the
production of oxidized bacterial cellulose with controllable
degradability was developed. For the rst time, BC membranes
were oxidized by NO –40% HNO solution, where the hydroxyl
groups of BC at C6 primarily reacted into carboxyl groups
during the oxidation. The degree of oxidation of the OBC
membranes increased as the oxidation time was extended and
eventually reached 32.54% aer oxidation for 40 h. The FTIR
and NMR spectra of the OBCs proved the formation of carboxyl
groups at the desirable C6 of in the glucopyranoside unit of the
BC chain. XRD analysis found that the oxidation process had
little impact on the crystal structure of BC, and the OBC samples
still retained the crystal structure of cellulose I. Crystallinity
6 W. Czaja, A. Krystynowicz, S. Bielecki and R. M. Brown Jr,
Microbial cellulose—the natural power to heal wounds,
Biomaterials, 2006, 27, 145–151.
7 W. K. Czaja, D. J. Young, M. Kawecki and R. M. Brown Jr, The
Future Prospects of Microbial Cellulose in Biomedical
Applications, Macromolecules, 2007, 8(1), 1–12.
8 P. C. Spencer, B. Schmidt, W. Samtleben, T. Bosch and
H. J. Gurland, Ex vivo model of heomodialysis membrane
biocompatibility, Trans. - Am. Soc. Artif. Intern. Organs,
1985, 31, 495–498.
9 D. Klemm, D. Schumann, U. Udhardt and S. Marsch,
Bacterial synthesized cellulose—articial blood vessels for
microsurgery, Prog. Polym. Sci., 2001, 26(9), 1561–1603.
2
3
studies revealed a slightly increasing trend which may result 10 J. A. Viljanto, Cellstick device for wound healing research, in
from the removal or rearrangement of the amorphous region
and the formation of more hydrogen bonds. The oxidation
process demonstrated a substantial toughening effect on the
Wound healing and skin physiology, ed. P. Altmeyer,
K. Hoffmann, S. el Grammal and J. Hutchinson, Springer,
Berlin, 1995, pp. 513–522.
nanocellulose brous network in evidence of signicant 11 J. D. Fontana, A. M. de Souza, C. K. Fontana, I. L. Torriani,
increases of tensile strength (44%), elongation at break (110%)
and toughness (121%) of the OBCs aer oxidation, despite a
minor decrease of Young's modulus. The degradation rate of
J. C. Moreschi, B. J. Gallotti, et al., Acetobacter cellulose
pellicle as a temporary skin substitute, Appl. Biochem.
Biotechnol., 1990, 24–25, 253–264.
OBC in PBS was signicantly improved, and increased as a 12 E. Entcheva, H. Bien, L. Yin, C.-Y. Chung, M. Farrell and
function of the degree of oxidation. Furthermore, with similar
levels of the oxidation degree, BC oxidized by NO –HNO solu-
Y. Kostov, Functional cardiac cell constructs on cellulose-
based scaffolding, Biomaterials, 2004, 25(26), 5753–5762.
2
3
tion in this work degraded relatively slower than BC oxidized by 13 M. Zaborowska, A. Bodin, H. Backdahl, J. Popp, A. Goldstein
NO
2
gas, showing a better controllability and a wider range of
and P. Gatenholm, Microporous bacterial cellulose as a
potential scaffold for bone regeneration, Acta Biomater.,
2010, 6, 2540–2547.
degradability. All OBC samples had a higher protein absorb-
ability than BC, and BC oxidized for 20 h with a degree of
oxidation of 11.57% absorbed the maximum amount BSA, 14 L. E. Millon and W. K. Wan, The polyvinyl alcohol–bacterial
conrming it good affinity for body uid and tissues.
cellulose system as a new nanocomposite for biomedical
1638 | RSC Adv., 2014, 4, 1630–1639
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
applications, J. Biomed. Mater. Res., Part B, 2006, 79(2), 245– 27 D. da Silva Perez, S. Montanari and M. R. Vignon, TEMPO-
2
53.
Mediated Oxidation of Cellulose III, Biomacromolecules,
2003, 4, 1417–1425.
28 A. Isogai and Y. Kato, Preparation of polyuronic acid from
cellulose by TEMPO-mediated oxidation, Cellulose, 1998, 5,
153–164.
1
1
5 A. Isogai, T. Saito and H. Fukuzumi, TEMPO-oxidized
cellulose, Nanobers, 2011, 3(1), 71–85.
6 S. Ifuku, M. Tsuji, M. Morimoto, H. Saimoto and H. Yano,
Synthesis of Silver Nanoparticles Templated by TEMPO-
Mediated Oxidized Bacterial Cellulose Nanobers, 29 J. B. Hanks, H. K. Kjaergard and D. A. Hollingsbee, A
®
Biomacromolecules, 2009, 10, 2714–2717.
Comparison of the Haemostatic Effect of Vivostat Patient-
®
1
7 J. Li, Y. Wan, L. Li, H. Liang and J. Wang, Preparation and
characterization of 2,3-dialdehyde bacterial cellulose for
potential biodegradable tissue engineering scaffolds,
Mater. Sci. Eng., C, 2009, 29, 1635–1642.
8 R. Benson, H. M. O'Neill, B. R. Evans, S. Hutchens,
C. P. Stephens and R. Hammonds, Development of
Bacterial Cellulose Nanocomposites, Mater. Res. Soc. Proc.,
Derived Fibrin Sealant with Oxidised Cellulose (Surgicel )
in Multiple Surgical Procedures, Eur. Surg. Res., 2003, 35,
439–444.
30 J. B. Sharma, M. Malhotra and P. Pundir, Laparoscopic
oxidized cellulose (Surgicel) application for small
uterine perforations, Int. J. Gynecol. Obstet., 2003, 83,
271–275.
1
2011, 1312, mrsf10-1312-ii06-09, DOI: 10.1557/opl.2011.112. 31 V. V. Klechkovskaya, et al., Structure of cellulose
19 X. Zhang, X. Li, H. Fan and X. Liu, Crystallinity and Thermal
Decomposition of Dialdehyde Celluloses from Bacterial
Cellulose, Key Eng. Mater., 2007, 330–332, 1289–1292.
20 T. Saito, S. Kimura, Y. Nishiyama and A. Isogai, Cellulose
Nanobers Prepared by TEMPO-Mediated Oxidation of
Native Cellulose, Biomacromolecules, 2007, 8, 2485–2491.
21 R. A. F. Clarke, Overview and general consideration of wound
Acetobacter xylinum, Crystallogr. Rep., 2003, 48(5), 755–
762.
32 C. R. Rambo, D. O. S. Recouvreux, C. A. Carminatti,
A. K. Pitlovanciv, R. V. Ant ˆo nio and L. M. Porto, Template
assisted synthesis of porous nanobrous cellulose
membranes for tissue engineering, Mater. Sci. Eng., C,
2008, 28(4), 549–554.
repair, in The Molecular and Cell Biology of Wound Repair, ed. 33 M. Roman and W. T. Winter, Effect of sulfate groups from
F. A. F. Clarke and P. M. Henson, Plenum Publishing, New
York, 1996, pp. 3–23.
2 Y. Marechal and H. Chanzy, The hydrogen bond network in
sulfuric acid hydrolysis on the thermal degradation
behavior of bacterial cellulose, Biomacromolecules, 2004, 5,
1671–1677.
2
2
Ib cellulose as observed by infrared spectrometry, J. Mol. 34 H. S. Barud, C. A. Ribeiro, M. S. Crespi, et al., Thermal
Struct., 2000, 523, 183–196.
characterization
of
bacterial
cellulose–phosphate
3 M. Kacurakova, A. C. Smith, M. J. Gidley, et al., Molecular
interactions in bacterial cellulose nanocomposites studied
composite membranes, J. Therm. Anal. Calorim., 2007,
87(3), 815–818.
by 1D FT-IR and dynamic 2D FT-IR spectroscopy, 35 H. S. Barud, A. M. de Araujo Junior, D. B. Santos, et al.,
Carbohydr. Res., 2002, 337, 1145–1153.
Thermal behavior of cellulose acetate produced from
homogeneous acetylation of bacterial cellulose,
Thermochim. Acta, 2008, 471, 61–69.
2
2
4 Q. G. Fan, D. M. Lewis and K. N. Tapley, Characterization of
Cellulose Aldehyde Using Fourier Transform Infrared
Spectroscopy, J. Appl. Polym. Sci., 2001, 82(5), 1195–1202.
5 S. Peng, Y. Zheng, J. Wu, Y. Wu, Y. Ma, W. Song and T. Xi,
Preparation and characterization of degradable oxidized
bacterial cellulose reacted with nitrogen dioxide, Polym.
Bull., 2012, 68, 415.
36 J. George, K. V. Ramana, A. S. Bawa and Siddaramaiah,
Bacterial cellulose nanocrystals exhibiting high thermal
stability and their polymer nanocomposites, Int. J. Biol.
Macromol., 2011, 48, 50–57.
37 S.
Ekici,
Intelligent
poly(N-isopropylacrylamide)-
2
6 U.-J. Kim, S. Kuga, M. Wada, T. Okano and T. Kondo,
carboxymethyl cellulose full interpenetrating polymeric
networks for protein adsorption studies, J. Mater. Sci.,
2011, 46, 2843–2850.
Periodate
Oxidation
of
Crystalline
Cellulose,
Biomacromolecules, 2000, 1, 488–492.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 1630–1639 | 1639