International Journal of

Molecular Sciences

Article
Preparation of Pure and Stable Chitosan Nanofibers
by Electrospinning in the Presence of
Poly(ethylene oxide)
Solomon Mengistu Lemma 1 , Frédéric Bossard 1 and Marguerite Rinaudo 2, *
1 Laboratoire Rhéologie et Procédés (LRP), University of Grenoble Alpes-CNRS, 38000 Grenoble, France;
solomonmengistu@gmail.com (S.M.L.); frederic.bossard@ujf-grenoble.fr (F.B.)
2 Biomaterials Applications, 6 Rue Lesdiguières, 38000 Grenoble, France
* Correspondence: marguerite.rinaudo@sfr.fr; Tel.: +33-61143-4806

Academic Editors: Hitoshi Sashiwa and Shinsuke Ifuku

Received: 22 September 2016; Accepted: 19 October 2016; Published: 26 October 2016

Abstract: Electrospinning was employed to obtain chitosan nanofibers from blends of chitosans (CS)
and poly(ethylene oxide) (PEO). Blends of chitosan (MW (weight-average molecular weight) =
102 kg/mol) and PEO (M (molecular weight) = 1000 kg/mol) were selected to optimize the
electrospinning process parameters. The PEO powder was solubilized into chitosan solution
at different weight ratios in 0.5 M acetic acid. The physicochemical changes of the nanofibers
were determined by scanning electron microscopy (SEM), swelling capacity, and nuclear magnetic
resonance (NMR) spectroscopy. For stabilization, the produced nanofibers were neutralized with
K2 CO3 in water or 70% ethanol/30% water as solvent. Subsequently, repeated washings with pure
water were performed to extract PEO, potassium acetate and carbonate salts formed in the course
of chitosan nanofiber purification. The increase of PEO content in the blend from 20 to 40 w%
exhibited bead-free fibers with average diameters 85 ± 19 and 147 ± 28 nm, respectively. Their NMR
analysis proved that PEO and the salts were nearly completely removed from the nanostructure
of chitosan, demonstrating that the adopted strategy is successful for producing pure chitosan
nanofibers. In addition, the nanofibers obtained after neutralization in ethanol-aqueous solution
has better structural stability, at least for six months in aqueous solutions (phosphate buffer (PBS)
or water).

Keywords: electrospinning; chitosan; PEO; nanofiber; stability

1. Introduction
Researchers have been working on the processing of chitosan to produce new biomaterials,
especially developed for biomedical applications [1–10]. The advantage of this natural polymer is
that it becomes water soluble in acidic conditions due to –NH2 protonation as soon as its degree of
acetylation is lower than 0.5. Then, processing of chitosan is relatively easy and it can be used as a
fiber, nanofiber, film, capsule, bead, sponge, gel, powder or tablet, based on its insolubility in neutral
medium (i.e., when chitosan is in the –NH2 form). Chitosan is often difficult to characterize but its main
characteristics are its weight-average molecular weight (MW ) and the average degree of acetylation
(DA) as discussed previously [11]. It is an interesting biodegradable and biocompatible polymer with
antibacterial and antifungal properties often described in the literature [12]. In addition, chitosan is
stabilized by H-bond network in solid state, providing good mechanical properties under film or fiber
materials. Biomedical applications are the main applications proposed for chitosan: in drug delivery,
as a gene delivery vehicle, for encapsulation of sensitive drugs, in medical textiles, in guided bone
regeneration, as a scaffold for nerve tissue regeneration, and in wound healing. The nanofibers are also

Int. J. Mol. Sci. 2016, 17, 1790; doi:10.3390/ijms17111790 www.mdpi.com/journal/ijms

Int. J. Mol. Sci. 2016, 17, 1790 2 of 16

used under membranes for fine filtration, and for adsorption of toxic metals [13]. The advantage of the
nanofibrous filter media is their high surface area per unit of mass and strong mechanical strength [14].
A wide range of methods has been used to produce chitosan-based nanofibrous materials, typically
electrospinning which has received a lot of attention in several scientific studies [1,15]. Electrospinning
of chitosan has been improved, optimizing the affecting factors such as the solution properties (viscosity,
surface tension and conductivity) [16–18], processing parameters and ambient conditions [16,19–22].
Chitosan has already been electrospun using different acid-based solvents [23,24]. For instance, acetic
acid used at concentration from 90% down to 1% (or 0.17 M) gives good fibers [16,25]. It is important
to mention that pure chitosan fibers have rarely been obtained in acetic acid at 90% and 70% [19], but
mainly with trifluoroacetic acid (TFA) [14,26] or mixture of TFA and dichloromethane [23,27]. The use
of ionic liquid was also mentioned with 1,1,1,3,3,3-hexafluoro-2-propanol [28]. More generally, it
has been shown that fine nanofibers can be obtained in chitosan blends with poly (vinyl alcohol)
(PVA) [23,25,29,30], gelatin or collagen [2,7,31–34], silk fibroin [5], polycaprolactone [35] but mainly
with poly(ethylene oxide) (PEO) [6,17,18,36–41]. Usually, to generate composite fibers the spinning
solutions are obtained by mixing the two polymers solutions prepared separately in the same solvent.
The processing of chitosan blends occurs by mono or coextrusion [20], and core-shell structured
PEO-chitosan nanofibers are also obtained, leading to a hollow nanofiber by removing PEO after
washing with water [42]. Interesting, shape-memory behavior was reported with relatively high
PEO content [40]. Good spinnability of PEO is recognized, and interaction with chitosan favors the
processing [17] in addition to its low toxicity [43].
The chitosan molecular weight from which nanofibers have been obtained varies from
85 kg/mol [42] up to 570 kg/mol [18,27,39], nevertheless the real molecular weight is often difficult
to obtain precisely due to the presence of aggregates [11]. Most generally moderate molecular
weights in the range of 100 to 200 kg/mol, used in different solvents. The concentration used for
electrospinning is directly related to the chitosan molecular weight which imposes the solution viscosity.
The concentration adopted is in the range of 20 to 50 g/L in different solvents [19,23,24,29,38]. It has
been demonstrated that low MW (as 30 kg/mol) or higher MW (as 398 kg/mol) are not able to give
good fibers in 90% acetic acid [24].
Due to swelling of the chitosan blend, crosslinking was proposed using glutaraldehyde [25,26]
or genipin [44]. Nevertheless, there are no very promising results after those chemical reactions.
Therefore, another more direct way should be modification of the intrinsic properties of chitosan,
which becomes insoluble in the –NH2 form. Then, due to the H-bond network established between
chitosan chains, no additional chemical reaction is needed. To obtain stabilization of the chitosan
nanofibers in aqueous medium, especially for biomedical applications, it is necessary to prevent
solubilization of chitosan due to its net charge under its salt form. The stability is usually tested by the
degree of swelling in phospahe buffer (PBS) or study of the morphology of fibers by scanning electron
microscopy (SEM). This problem is not often discussed in the literature. The use of acid with a low
boiling temperature (such as acetic acid) allows extraction of excess acid and water constituting the
solvent left after the electrospinning in ambient conditions. Then, a neutralization step is important
to obtain the insoluble chitosan. For this purpose, it was proposed that the nanofiber membrane be
immersed in 5 M NaOH or 5 M Na2 CO3 aqueous solutions for 3 h at ambient conditions [27] or in 1 M
K2 CO3 for 3 h at 25 ◦ C [14].
With respect to the applications of these chitosan materials obtained by electrospinning, it must
be pointed out that they are important for cell cultures [5,37], tissue engineering [6], and also for ion
adsorption [14,39,41] or water treatment [15]. It is also necessary to mention that not only electrospun
chitin and chitosan are produced but also chitin derivatives [24,45,46].
In our work, the electrospinning of chitosan-based nanofibers was developed in order to obtain
a higher yield of chitosan. For that purpose, two different techniques were used. Firstly, solutions
of PEO and solutions of chitosan were mixed to produce nanofibers. Secondly, powdered PEO was
added into chitosan solutions. Due to toxicity of the solvents able to be used, the acetic acid in
Int. J. Mol. Sci. 2016, 17, 1790 3 of 16

absence of any additive was chosen at the concentration of 0.5 M corresponding to maximum of
chitosan solubility [47]. The development of chitosan nanofibers without toxic solvent allows for
production of a better candidate for biomedical, food and pharmaceutical applications. The change
of theInt. J. Mol. Sci. 2016, 17, 1790
molecular weight and weight ratio of PEO in the chitosan blend were investigated and 3 of 16
related
with of
theaelectrospun nanofiber thickness. The primary aim was achieved to develop
better candidate for biomedical, food and pharmaceutical applications. The change of the
stable and pure
chitosan nanofibers
molecular weightwithout
and weightPEO andofremaining
ratio PEO in the salts. Furthermore,
chitosan the influence
blend were investigated andof neutralization
related with
step on the morphology stability of these fibers was also covered.
the electrospun nanofiber thickness. The primary aim was achieved to develop stable and pure
chitosan nanofibers without PEO and remaining salts. Furthermore, the influence of neutralization
2. Results
step onand Discussionsstability of these fibers was also covered.
the morphology
Production of regular electrospun nanofibers from pure chitosan solutions is known to be
2. Results and Discussions
challenging. Due to this, chitosan/PEO (CS/PEO) blends were used to enhance the spinnability
Production
of different of regular
MW chitosans electrospun
in a common nanofibers
solvent. The from pure of
influence chitosan solutions
PEO was is known
examined in thetorange
be of
challenging. Due to this, chitosan/PEO (CS/PEO) blends were used to enhance the spinnability
high amount of chitosan compared with PEO with the objective of producing pure chitosan fibers. of
different MW chitosans in a common solvent. The influence of PEO was examined in the range of high
amount of chitosan
2.1. Chitosan/PEO compared
(CS/PEO) with PEOBased
Compatibility with the
on objective of producing pure chitosan fibers.
Solution Viscosity
First of all, the solvent
2.1. Chitosan/PEO (CS/PEO)adopted was fixed
Compatibility Based aton
0.5Solution
M acetic acid after testing 90% acetic acid which is
Viscosity
too volatile. Electrospinning of pure chitosan solution leads
First of all, the solvent adopted was fixed at 0.5 M acetic acid to the formation
after of beads
testing 90% instead
acetic acid whichof fine
nanofibers.
is too volatile. Electrospinning of pure chitosan solution leads to the formation of beads instead of mix
Then, taking into account data from literature [10,17,18,35,36,42], it was decided to
chitosan
fine with PEO, allowing
nanofibers. forinto
Then, taking better electrospinning.
account data from literature [10,17,18,35,36,42], it was decided to
In
mixparallel,
chitosan to test
with theallowing
PEO, compatibility of electrospinning.
for better these two polymers, the viscosities of mixed solutions
In parallel, to test the compatibility
prepared at 5% (w/v) were determined. All the of these two polymers,
solutions testedthe
wereviscosities of mixed
Newtonian solutions
at low shear rate.
prepared at 5% (w/v) were determined. All the solutions tested were − Newtonian
1
The main data are given in Figure 1 where the viscosities at 0.5 s were plotted as a function at low shear rate. of
The main data are given in Figure 1 where the viscosities at 0.5 s−1 were plotted as a function of
CS/PEO ratio. The total polymer concentration remains constant and equal to 5% w/v. These results
CS/PEO ratio. The total polymer concentration remains constant and equal to 5% w/v. These results
confirm the synergistic interaction between the two polymers in the chosen solvent used as mentioned
confirm the synergistic interaction between the two polymers in the chosen solvent used as
in the literature [17,48,49]. An optimum viscosity value was obtained around a 50/50 (v/v) ratio but
mentioned in the literature [17,48,49]. An optimum viscosity value was obtained around a 50/50 (v/v)
thereratio
was but
no there
significant
was nochange in the
significant viscoelastic
change behaviorbehavior
in the viscoelastic of the solutions.
of the solutions.

Figure 1. Variation of the viscosity of chitosan CS-102/ PEO (M = 1000 kg/mol) mixtures in 0.5 M acetic
Figure 1. Variation of the viscosity of chitosan CS-102/ PEO (M = 1000 kg/mol) mixtures in 0.5 M
acid at 0.5 s−1 as a function of weight fraction of PEO (Xv). Total polymer concentration 5% w/v. PEO:
acetic acid at 0.5 s−1 as a function of weight fraction of PEO (Xv ). Total polymer concentration 5% w/v.
poly(ethylene oxide); CS: Chitosan.
PEO: poly(ethylene oxide); CS: Chitosan.
A second technique used to prepare the chitosan/PEO solution consisted of adding solid PEO
into
A 5% (w/v)
second chitosan
technique usedsolution. As the
to prepare shown in Figure 2,solution
chitosan/PEO the viscosity increased
consisted progressively,
of adding solid PEO into
depending
5% (w/v) on the
chitosan concentration
solution. As shown and in
molar mass
Figure of PEO
2, the used. increased
viscosity In these conditions, the chitosan
progressively, depending
concentration remained constant (5% w/v). Up to the ratio 60/40, G’ < G’’ in the
on the concentration and molar mass of PEO used. In these conditions, the chitosan concentration entire range of
frequency was tested but the solutions become non-Newtonian. These data confirm the cooperative
remained constant (5% w/v). Up to the ratio 60/40, G’ < G” in the entire range of frequency was tested
interaction between chitosan and PEO especially when the experimental data from Figure 2, curve c
but the solutions become non-Newtonian. These data confirm the cooperative interaction between
is compared to the additivity calculated in Figure 2, curve b. All these solutions were electrospun
chitosan andany
without PEO especially
difficulty, when
except purethe experimental
chitosan datablended
and chitosan from Figure 2, curve
with 10% PEO. c is compared to the
additivity calculated in Figure 2, curve b. All these solutions were electrospun without any difficulty,
except pure chitosan and chitosan blended with 10% PEO.
 Int. J. Mol. Sci. 2016, 17, 1790 4 of 16
Int. J. Mol. Sci. 2016, 17, 1790 4 of 16
Int. J. Mol. Sci. 2016, 17, 1790 4 of 16

Figure 2. Variation of the viscosity at 0.5 s−1 of added PEO (M = 1000 kg/mol) expressed in percent of
Figure2.2. Variation
Figure Variationof ofthe
the viscosity
viscosity at 0.5 ss−−11of
at 0.5 ofadded
addedPEO
PEO(M(M==1000
1000kg/mol)
kg/mol) expressed
expressed inin
percent
percentof
total polymer concentration in the mixtures. (a) PEO solution in 0.5 M acetic acid ; (b) additivity
total
of polymer
total polymer concentration
concentration ininthe
themixtures.
mixtures.(a) (a)PEO
PEOsolution
solutioninin0.5
0.5MM acetic acid ;; (b)
acetic acid (b) additivity
additivity
calculated for each values of “a” added into pure chitosan CS-102  and (c) blends chitosan
calculated
calculated foreach
for each values
values of added
of “a” “a” added
into pure intochitosan
CS-102/PEO as a function of the weight fraction of PEO .
pure chitosan
CS-102 ? CS-102  andchitosan
and (c) blends (c) blends chitosan
CS-102/PEO
as a function of
CS-102/PEO as the weight of
a function fraction of PEO
the weight .
fraction of PEO .○
2.2. Polymers Concentration and Molecular Weight Influences
2.2. Polymers
2.2. Polymers Concentration
Concentration and
and Molecular
Molecular Weight
WeightInfluences
Influences
A number of experiments were carried out to determine the optimum processing conditions and
blend
A ratio ofof
A number ofCS/PEO in orderwere
experiments
experiments to getcarried
were nanofibrous
carriedout tomatrix.
out First, PEO
determine
to determine thethe(300
optimumkg/mol)
optimum atprocessing
5% (w/w)
processing solution
conditions
conditionsand
blend
and in 0.5 M acetic
ratioratio
blend of CS/PEO acid
of CS/PEOwas electrospun
in order to getto
in order to optimize
nanofibrous the production
matrix.
get nanofibrous of
First, PEO
matrix. fibers as
First,(300
PEO shown
kg/mol) in the Table
at 5% (w/w)
(300 kg/mol) 1.
at 5%The
solution
(w/w)
in microscope
0.5 M acetic observation
acid was confirms that
electrospun to PEO can the
optimize giveproduction
fibers at this of molecular
fibers as weight in
shown with
theadopted
Table 1. The
solution in 0.5 M acetic acid was electrospun to optimize the production of fibers as shown in the
electrospinning parameters. In the same processing conditions, chitosan (102 kg/mol) solution at 5%
microscope
Table 1. Theobservation
microscopeconfirms
observationthat confirms
PEO can that
givePEO fiberscanat give
this molecular
fibers at thisweight with adopted
molecular weight
(w/w) in 0.5 M acetic acid was also tested to form fibers. Unlike PEO, electrospinning of this chitosan
electrospinning
with parameters. Inparameters.
adoptedgenerated
electrospinning the same processing conditions, chitosan (102 kg/mol) solution at 5%
solution mainly beads (Table 1).In the same processing conditions, chitosan (102 kg/mol)
solution at 5% (w/w) in 0.5 M acetic tested
(w/w) in 0.5 M acetic acid was also acid wasto form fibers.toUnlike
also tested PEO, electrospinning
form fibers. of this chitosan
Unlike PEO, electrospinning of
solution
this generated
chitosan solution
Table mainly beads
generated
1. Electrospun (Tablebeads
pure mainly
chitosan, 1). (Table
PEO (300 1). and blends of CS/PEO at total polymer
kg/mol)
concentration (50 mg/mL) at 20 kV applied voltage, 50 µL/h pump rate and 15 cm collection.
Table1.1. Electrospun
Table Electrospun purepure chitosan,
chitosan, PEO
PEO (300
(300 kg/mol)
kg/mol) and
and blends
blends of
of CS/PEO
CS/PEO atat total
total polymer
polymer
CS (mg/mL) PEO (mg/mL) CS MW (kg/mol) CS/PEO (w/w)% β Electrospun Product
concentration (50 mg/mL)
concentration (50 mg/mL) at 20 kV applied voltage, 50 µL/h pump rate and 15 cm collection.
µL/h pump rate and 15 cm collection.
0 50 0 0/100 Fibers
CS (mg/mL)50 PEO (mg/mL)0 CS MW102 (kg/mol) CS/PEO 0/100(w/w)% β Electrospun
Beads a Product
CS (mg/mL) PEO (mg/mL) CS M W102
(kg/mol) CS/PEO (w/w)% β Electrospun Product
0 45 50 5 0 90/10
0/100 BeadsFibers
0 40 50 10 102
0 80/20
0/100 Fibers + beads
Fibers
50 0 102 0/100 Beadsba a
50 35 0 15 102
102 0/100
70/30 Beads
Fibers + beads
4545 30 55 20 102
102
102
90/10
90/10
60/40
Beads
Beads
Fibers + beads
4040 25 101025 102
102
102 80/20
80/20
50/50 Fibers
Fibers + +beads
Fibers + beads beads
35β35 1515 102
102 70/30a
70/30 Fibers
Fibers + + beads
beads bb
Beads and fibers examined using Leica DM LM microscope (50×); elongated bead obtained at
3030 20 102 60/40 Fibers+ +beads
beads
30 µL/h and b fibers 20 102 obtained at 20 µL/h.
having beads along them 60/40 Fibers
25 25 102 50/50 Fibers + beads
25 25 102 50/50 Fibers + beads
βBeads and fibers
Therefore, examinedofusing
blending Leica DM LM microscope ); aaelongated
(50×at bead polymer
obtained atconcentration
30 µL/h and
β
bBeads and fibers examinedthe twoLeica
using polymers
DM LMsolutions
microscope total
(50×); fixed
a elongated bead obtained at
fibers having beads along them obtained at 20 µL/h.
(5% w/w) was selected as an alternative option to obtain composite fibers. In addition, the pump rate
30 µL/h and b fibers having beads along them obtained at 20 µL/h.
was also kept at 50 µL/h in all electrospun samples. Exceptionally, blended CS/PEO (90/10) and
CS/PEO (70/30)
Therefore,
Therefore,blending
formed the
blendingof of
elongated
twotwo
the
beads and
polymers fibers at a30total
solutions
polymers
and 20 µL/h
solutions at fixed
pump rate, respectively.
polymer
a total fixed concentration (5% w/w)
polymer concentration
was However,asTable
selected an 1 shows that
alternative optionregardless
to obtainof composite
the blend fibers.
ratio and
In change inthe
addition, thepump
corresponding
rate was also
(5% w/w) was selected as an alternative option to obtain composite fibers. In addition, the pump rate
kept processing
at 50 kept parameters,
µL/hatin50allµL/h well-structured
electrospun nanofibers were
samples. Exceptionally, not obtained. One
blended CS/PEO of the reasons
(90/10) might
and be
CS/PEO
was that
also in all electrospun samples. Exceptionally, blended
mixing the two solutions causes a decrease in the final chitosan and PEO concentrations. CS/PEO (90/10) and
(70/30)
CS/PEOformed
(70/30)elongated
formed beads and fibers
elongated beadsatand
30 and 20 µL/h
fibers at 30pump
and rate,
20 respectively.
µL/h pump However,
rate, Table 1
respectively.
Thus, electrospinning of blend CS/PEO at 5% w/w from chitosan (102 kg/mol) and PEO
shows
However,that Table
regardless of thethat
1 shows blend ratio andofchange
regardless the blend in the corresponding
ratio and processing parameters,
(300 kg/mol) solutions in acetic acid was not successful in producing beadchange in For
free fibers. thethis
corresponding
reason,
well-structured
processing
powdered
nanofibers
parameters,
PEO (with five
were not obtained.
well-structured nanofibers
different molecular
One
weights)
of
were theadded
werenot
reasons
obtained.might be
of that
Onechitosan
to 5% (w/w)
mixing might
the reasons
(102 kg/mol)
the twobe
solutions causes
that solutions
mixing the a decrease
to control in the
CS/PEO causes
two solutions final chitosan
ratios. The and
additionin
a decrease PEO
of the
powderedconcentrations.
PEO helps
final chitosan andto PEO
avoidconcentrations.
the dilution effect
Thus,
of electrospinning
the final
Thus, of
of blend
chitosan concentration
electrospinning CS/PEO
blendand allowsat
CS/PEO 5%5% w/w
producing
at from
from chitosan
w/wnanofibers (102
as reported
chitosan (102 inkg/mol)
Table 2.and
kg/mol) The PEO
and PEO
(300
(300 kg/mol) solutionsininacetic
kg/mol) solutions acetic acid
acid was
was notnot successful
successful in producing
in producing beadbead free fibers.
free fibers. For thisFor this
reason,
reason, powdered PEO (with five different molecular weights) were added to 5% (w/w)
powdered PEO (with five different molecular weights) were added to 5% (w/w) chitosan (102 kg/mol)chitosan
(102 kg/mol)
solutions solutions
to control to control
CS/PEO CS/PEO
ratios. ratios.of
The addition The additionPEO
powdered of powdered PEO helps
helps to avoid to avoid
the dilution the
effect
of the final chitosan concentration and allows producing nanofibers as reported in Table 2. The
Int. J. Mol. Sci. 2016, 17, 1790 5 of 16

dilution effect of the final chitosan concentration and allows producing nanofibers as reported in
Table 2. The processing parameters such as pump rate, collection distance and applied voltage were
kept the same as those adopted for data given in Table 1. Nevertheless, in some cases, the collection
distance and pump rate were changed to optimize the processing parameters. Microscope images of
these different blended samples showed that the addition of powdered PEO favors the formation of
fibers in contrary to those obtained in the conditions adopted in Table 1. In general, the addition of
powdered PEO into chitosan solutions to produce electrospun chitosan-PEO fiber matrix has never
before been reported.

Table 2. Composition of electrospun solutions obtained with different molecular weights PEO added
into chitosan CS (MW = 102 kg/mol) in 0.5 M acetic acid solution at 20 kV applied voltage.

CS PEO PEO M CS/PEO Pump Rate β Electrospun

(mg/mL) (mg/mL) (kg/mol) (w/w)% (µL/h) Products
50 12.0 100 80/20 20, 30, 50 & 80 Beads a
50 12.0 300 80/20 50 Beads
50 12.0 1000 80/20 50 & 100 Fibers
50 12.0 5000 80/20 20 Fibers b
50 12.0 8000 80/20 80 Fibers
40 5.6 8000 90/10 50 Fibers
40 5.6 5000 90/10 10 Fibers + beads
50 10.0 300 80/20 20 Fibers + beads c
50 20.0 300 70/30 20 Fibers
50 30.0 300 60/40 50 Fibers
50 40.0 300 55/45 50 Fibers
β Beads and fibers examined using Leica DMLM microscopes (50×). In all cases the collection distance is 15 cm,
except “a” used both 15 & 10 cm, b = 10 cm and c = 11 cm.

Table 2 firstly allowed the role of PEO molecular weight at a CS/PEO ratio 80/20 to be shown.
At lower PEO molecular weights no fibers were obtained. On the hand, fibers were obtained
with chitosan (102 kg/mol) blended with PEO, with molecular weights equal to or higher than
M 1000 kg/mol. This M PEO was selected for further production of chitosan-PEO matrices (shown in
Table 3). In addition, it is observed that, when the M of PEO is larger (M 8000 kg/mol), fibers were
obtained at an even lower PEO content (at CS/PEO ratio 90/10). From complementary results shown
in Table 2, the influence of PEO (M 300 kg/mol) content was demonstrated: bead free fibers were
obtained for CS/PEO ratios lower than 80/20 in relation with the good spinnability of PEO. Finally, the
influence of the chitosan molecular weight was examined. Results obtained when powdered PEO (M
1000 kg/mol) was added in chitosan solutions (CS-102 at 5% and CS-500 at 1.4%) are given in Table 3.

Table 3. Electrospun blends of chitosans having different molecular weights with PEO (1000 kg/mol).

CS PEO CS-M W CS/PEO Pump Rate β Electrospun

(mg/mL) (mg/mL) (kg/mol) (w/w)% (µL/h) Products
35.8 4.2 102 90/10 20 Beads
32.0 8.0 102 80/20 50 Fibers + beads
28.6 11.4 102 70/30 50 a Fibers
13.5 1.6 500 * 90/10 50 c Fibers + beads
12.8 3.4 500 * 80/20 50 Fibers + beads
12.3 4.9 500 * 70/30 50 Fibers + beads
11.2 7.8 500 * 60/40 20 b Beads
* 1.4% chitosan solution. β Beads and fibers examined using Leica DMLM microscopes (50×). The collection
distance was 20 cm except a = 15 cm and b = 16 cm. In all cases an applied voltage of 20 kV was used, except
a = 15 kV, b = 25 kV and c = 16 kV.

Different composition ratios were covered as shown in Table 3. Overall, the results confirm that
the blend with chitosan (102 kg/mol) gave better fibers compared to the higher molecular weight
chitosan (500 kg/mol) at the same CS/PEO ratio. Particularly, using the higher MW chitosan, it was
Int. J. Mol. Sci. 2016, 17, 1790 6 of 16

impossible to increase the concentration above 1.4% w/w. This is due to the increased viscosity of
the blend when PEO dissolved in this chitosan solution. These results indicate that chitosan with
Int. J. Mol. Sci. 2016, 17, 1790 6 of 16
moderate MW favors fiber production, allowing an increase chitosan content. Similar conclusions on
chitosan MW influence
Different were obtained
composition previously
ratios were covered asfor chitosan
shown dissolved
in Table 3. Overall,in
theTFA-DCM system
results confirm that and in
90% acetic
the acid
blendrespectively
with chitosan[9,24]. (102 kg/mol) gave better fibers compared to the higher molecular weight
chitosan
Taking into(500 kg/mol)
account our at experimental
the same CS/PEO ratio.
data, 5%Particularly, using solutions
(w/v) chitosan the higher (102
MW chitosan, it was
kg/mol)/powdered
impossible to increase the concentration above 1.4% w/w. This is due to the increased viscosity of
PEO (1000 kg/mol) were selected for preparing the chitosan/PEO ratios 90/10, 80/20, 70/30 and
the blend when PEO dissolved in this chitosan solution. These results indicate that chitosan with
60/40 (w/w). These blended solutions were used to produce electrospun nanofibers in optimal
moderate MW favors fiber production, allowing an increase chitosan content. Similar conclusions on
conditions and
chitosan M toWperform
influence further physicochemical
were obtained previously foranalyses.
chitosan dissolved in TFA-DCM system and in
90% acetic acid respectively [9,24].
2.3. Morphology andinto
Taking Stability
accountofour
theexperimental
Chitosan Nanofibers
data, 5% (w/v) chitosan solutions (102 kg/mol)/powdered
PEO (1000 kg/mol) were selected for preparing the chitosan/PEO ratios 90/10, 80/20, 70/30 and 60/40
2.3.1. Electrospun
(w/w). TheseChitosan/PEO
blended solutionsNanofibers
were used to produce electrospun nanofibers in optimal conditions
and to perform further physicochemical analyses.
A selected chitosan 5% (w/w) solution/powdered PEO, with molecular weight of 102 and
1000 kg/mol, respectively,
2.3. Morphology were
and Stability used
of the to obtain
Chitosan stable nanofibrous chitosan as shown in Figure 3.
Nanofibers
The processing parameters of electrospinning were optimized at pump rate 20 µL/h, applied voltage
of 20 kV2.3.1.
andElectrospun Chitosan/PEO
tip to collector distanceNanofibers
of 13 cm. Furthermore, the conductivity of the blends decreased
with respectAtoselected
the reduction
chitosan of 5%weight ratio of protonated
(w/w) solution/powdered PEO, chitosan in CS/PEO
with molecular weightsolutions.
of 102 andWith
1000 respect
kg/mol, respectively,
to the increment were used PEO
of 10% powdered to obtain
in the stable nanofibrous
blends of CS/PEO chitosan
from as shown
10% toin40%, Figure 3. The
these solutions
processing parameters of electrospinning were optimized at pump rate
have conductivity values of 7.21 ± 0.01, 6.95 ± 0.02, 6.53 ± 0.01 and 6.22 ± 0.03 mS/cm, respectively.20 µL/h, applied voltage of
20 kV and tip to collector distance of 13 cm. Furthermore, the conductivity of the blends decreased
This decrement of the conductivity leads to an increase in the average fiber diameter (AFD) due to
with respect to the reduction of weight ratio of protonated chitosan in CS/PEO solutions. With respect
lower stretching of theofsolution
to the increment at constant
10% powdered PEO in strong electric
the blends field [17].
of CS/PEO from 10% to 40%, these solutions
As have
can be seen in Figure
conductivity values 3a, the ±results
of 7.21 of ±the
0.01, 6.95 blended
0.02, 6.53 ± 0.01CS/PEO
and 6.22(90/10) exhibitrespectively.
± 0.03 mS/cm, more nano-beads
with average diameters
This decrement of the (244 ± 61)leads
ofconductivity nm. toThese beads
an increase in are connected
the average fiber with
diameterfibers having
(AFD) due toAFD of
lower stretching of the solution at constant strong electric field [17].
46 ± 13 nm. This result led to the assumption that the beads are formed from chitosan and the
connecting As can be seen in Figure 3a, the results of the blended CS/PEO (90/10) exhibit more nano-beads
fibers from PEO. It is evident that 10% of PEO in the blend did not prevent chitosan
with average diameters of (244 ± 61) nm. These beads are connected with fibers having AFD of
from forming bead-free fibers and stable Taylor cone. This result is consistent with results in other
46 ± 13 nm. This result led to the assumption that the beads are formed from chitosan and the
studies connecting
using different solvents
fibers from PEO. It[9,17,37].
is evident Accordingly,
that 10% of PEOpowdered
in the blend PEO did not content
preventincreased
chitosan from to 20% in
the blend, favoring
forming the optimized
bead-free processing
fibers and stable parameters
Taylor cone. This result and producing
is consistent withnano-scale fibers
results in other of chitosan.
studies
At 80/20using
ratiodifferent solventswell-structured
of CS/PEO, [9,17,37]. Accordingly, fine powdered
nanofibers PEO content
were increased
formed withtoaverage
20% in the blend,
fiber diameter
favoring the optimized processing parameters and producing nano-scale
(AFD) of 85 ± 19 nm, together with very few elongated beads (Figure 3b). In the case of 70/30 and fibers of chitosan. At 80/20
ratio of CS/PEO, well-structured fine nanofibers were formed with average fiber diameter (AFD) of
60/40, the smooth nanofibers free of beads were obtained; determined AFD was 146 ± 31 nm in
85 ± 19 nm, together with very few elongated beads (Figure 3b). In the case of 70/30 and 60/40, the
Figure 3c
smooth 147 ± 28 free
and nanofibers nm of inbeads
Figure were3d,obtained;
respectively. These
determined AFD results
was 146 clearly
± 31 nm revealed
in Figurethat bead free
3c and
nanofibers
147 ±can be produced
28 nm in Figure 3d,when the content
respectively. These of chitosan
results clearlyprepared at 5%
revealed that beadinitial concentration
free nanofibers can beis equal
to or less than 80%
produced when inthe
thecontent
blend.of In addition,
chitosan preparedthe at
AFD of theconcentration
5% initial blend at 70/30 is very
is equal close
to or less thanto 60/40.
80% in the blend. In addition, the AFD of the blend at 70/30 is very
However, the increased content of powdered PEO from 20% to 30% in the blends increased the AFD close to 60/40. However, the
by 41%.increased content of powdered PEO from 20% to 30% in the blends increased the AFD by 41%. This
This indicates that the increased content of chitosan together with optimal blending ratio of
indicates that the increased content of chitosan together with optimal blending ratio of powdered
powdered PEO has significant effect on nanofiber size.
PEO has significant effect on nanofiber size.

Figure 3. Cont.
Int. J. Mol. Sci. 2016, 17, 1790 7 of 16
Int. J. Mol. Sci. 2016, 17, 1790 7 of 16

Figure 3. Representative SEM images and fiber diameter distributions of the electrospun nanofibers
Figure 3. Representative
obtained from blendSEM images PEO
of powdered and into
fiberchitosan
diameter distributions
solutions (CS/PEO) of
at the electrospun
weight nanofibers
ratio (a) 90/10;
obtained(b)from
80/20;blend of and
(c) 70/30; powdered
(d) 60/40.PEO into chitosan
The insets solutions
show the high (CS/PEO)
magnification images.at The
weight ratio (a) 90/10;
fiber diameter
(b) 80/20; (c) 70/30;
distributions andare
of each (d)shown
60/40. The
on the insets
right. AFD:show thefiber
average high magnification
diameter; images.
SEM: scanning The fiber
electron
diametermicroscopy.
distributions of each are shown on the right. AFD: average fiber diameter; SEM: scanning
electron microscopy.
2.3.2. Stabilization of Chitosan Nanofibers
Electrospinning
2.3.2. Stabilization of CS/PEO
of Chitosan at ratio of 80/20 and 60/40 nanofibers were chosen to examine
Nanofibers
structural stability after repeated extraction of PEO. Specifically, the nanofibers from CS/PEO (80/20)
Electrospinning of CS/PEO
were studied in reason at ratio
of their high of of
content 80/20 and
chitosan in 60/40 nanofibers
the blend. were chosen
The lower content to at
of chitosan examine
thestability
structural ratio CS/PEO
after(60/40) was investigated
repeated extraction of also and compared
PEO. withthe
Specifically, CS/PEO (80/20).from
nanofibers To obtain
CS/PEOstable (80/20)
chitosan nanofibrous membranes, washing out PEO from CS/PEO nanofibers is necessary. However,
were studied in reason of their high content of chitosan in the blend. The lower content of chitosan
before total PEO extraction, neutralization of the chitosan fibers needs to be performed. For that
at the ratio CS/PEO (60/40) was investigated also and compared with CS/PEO (80/20). To obtain
purpose and following the literature, the nanofibers membranes were immersed in aqueous 1 M
stable chitosan
K2CO3 for two days with membranes,
nanofibrous exchange of thewashing out PEO
washing solution fourfrom
timesCS/PEO nanofibers
at 25 °C [14,24]. is necessary.
This allowed
However, before totalofPEO
for neutralizing extraction,
the chitosan nanofiberneutralization of the chitosan
membranes, becoming fibers
insoluble in needs
aqueous to beunder
medium performed.
For thatthe –NH2 form.
purpose In addition,
and following theit literature,
is known that many charged
the nanofibers polysaccharides
membranes are insolubleinin
were immersed aqueous
ethanol/water
1 M K2 CO for two mixtures.
days For exchange
with instance, it has
of beenwashing
the shown that in presence
solution fourof times
2% acetic
at 25 ◦C
acid, chitosan
[14,24]. This
3
is insoluble at ethanol yield over 40% [50]. Then, a new post-treatment was proposed, involving the
allowed for neutralizing of the chitosan nanofiber membranes, becoming insoluble in aqueous medium
neutralization of chitosan in presence of ethanol, a non-solvent of chitosan. Stabilization was
under the –NH2 form. In addition, it is known that many charged polysaccharides are insoluble in
ethanol/water mixtures. For instance, it has been shown that in presence of 2% acetic acid, chitosan
is insoluble at ethanol yield over 40% [50]. Then, a new post-treatment was proposed, involving
the neutralization of chitosan in presence of ethanol, a non-solvent of chitosan. Stabilization was
performed in 70/30 ethanol/aqueous 1 M K2 CO3 treatment to get the insoluble -NH2 form chitosan,
followed by water washing. Then, the two methods of neutralization were compared using K2 CO3
Int. J. Mol. Sci. 2016, 17, 1790 8 of 16
Int. J. Mol. Sci. 2016, 17, 1790 8 of 16
performed in 70/30 ethanol/aqueous 1 M K2CO3 treatment to get the insoluble -NH2 form chitosan,
followed by water washing. Then, the two methods of neutralization were compared using K2CO3 in
in 100%
100% water
water oror70%/30%
70%/30% ethanol-water
ethanol-water mixture.These
mixture. Thesetreatments
treatmentswere
werecarried
carriedout
out to
to check
check the
the
influence
influence of neutralization conditions on the stability of the chitosan nanofibers structure as depicted
of neutralization conditions on the stability of the chitosan nanofibers structure as depicted
in
in Figure
Figure 4.
4.

Figure 4. The SEM images of the membrane surface of chitosan/PEO at different ratios. (a) 60/40
Figure 4. The SEM images of the membrane surface of chitosan/PEO at different ratios. (a) 60/40
neutralized in water; (b) 60/40 neutralized in ethanol/water mixture; (c) 80/20 neutralized in water;
neutralized in water; (b) 60/40 neutralized in ethanol/water mixture; (c) 80/20 neutralized in water;
and (d) 80/20 neutralized in ethanol/water mixture.
and (d) 80/20 neutralized in ethanol/water mixture.
Int. J. Mol. Sci. 2016, 17, 1790 9 of 16

Figure 4 shows the SEM images of CS/PEO before and after immersion in both neutralizing
media. The nanofibrous membranes neutralized in water failed to preserve the chitosan nanofiber
structure due to partial dissolution when the initial fibers under the acetate ionic form were immersed
in 1 M K2 CO3 . This is evidenced by the existence of a film-like structure (Figure 4a,c). Typically,
in the SEM image, CS/PEO (60/40) lost its nano-structured fibers (Figure 4a) more than CS/PEO
(80/20) as shown in Figure 4c. This may be due to larger amount of PEO washed out from the
composite nanofibrous structure and partial swelling of protonated chitosan at first contact with
aqueous solution. On the contrary, SEM images analyses confirmed the stability of nanofibrous
structures after treatment with ethanol-water (70/30) as shown in Figure 4b,d. Generally, the AFD
determined after immersion in ethanol-water (70/30) was increased by around 30% for CS/PEO
(80/20) and 33% for CS/PEO (60/40) compared with the initial size (Figure 3b,d). This size increment
may be due to the swelling of the fibers. This is in good agreement with previous studies which
indicated that washing of chitosan membranes increased the size of the nanofibers by swelling without
change of fibrous morphology [37,51]. In this case, it is shown that the fiber morphology is perfectly
preserved and consists of pure chitosan as demonstrated by nuclear magnetic resonance (NMR).
To conclude, chitosan neutralization in ethanol-water mixture is suitable for preserving the nanofiber
structure as shown in Figure 4b,d. A similar conclusion was obtained previously when 3 M NaOH in
aqueous solution of methanol was used [51].

2.4. Characterization of Chitosan Nanofibers

The electrospun nanofibers obtained are composed of interacting PEO and chitosan in the acetate
salt form which are water soluble in acidic conditions. For this purpose, excess of acetic acid and water
may be easily separated after processing by evaporation in ambient conditions. When these fibers were
immersed in PBS buffer at pH = 7.4 or in water, the fibers were swollen and PEO was progressively
solubilized, modifying their properties. In our approach, to obtain pure chitosan fibers, the electrospun
fibers were immersed in K2 CO3 to neutralize chitosan and then in water to extract PEO, and potassium
acetate and carbonate salt formed. The proposed neutralization post-treatment was applied in presence
of ethanol to get pure chitosan nanofibers with preserved morphology (Figure 4b,d). After repeated
washing in water, the residual final fibers should be composed of pure chitosan in the –NH2 form.
The analysis of fibers and role of different steps of purification were tested by NMR spectroscopy.
Examples are given in the following section.

2.4.1. NMR Analysis and Composition of the Fibers

Nuclear magnetic resonance (NMR) spectroscopy was used to investigate chemical changes
of nanofibers of CS/PEO. This helps to identify the remaining PEO and acetic acid in the chitosan
nanofibers. In Figure 5, the NMR spectrum of a CS/PEO (80/20) blend as used for electrospinning
(Figure 5A) was compared with that of chitosan in the same solvent (Figure 5B). The integrals of the
main signals are given in the Figure. Considering the chitosan spectrum (Figure 5B), the following
signals are identified at 2.15 ppm corresponding to the –CH3 protons and at 3.13 ppm for the proton
in –C2 position. The ratio between these two integrals allows to determine the DA of chitosan [52].
With respect to the analysis of the mixture of CS/PEO solution prepared in acetic acid and added
D2 O (Figure 5A), the signal of –CH2 – from PEO at 3.73 ppm is superimposed with the signal of one
chitosan proton. Therefore, the quantitative determination of PEO content should be overestimated.
Nevertheless, this thin signal allows for a conclusion on the presence or absence of PEO in the
fibers analyzed after purification. In the two spectra, a large signal at 2.13 ppm (with two satellites)
allows us to determine the acetic acid concentration used for electrospinning when the exact chitosan
concentration is known.
Int. J. Mol. Sci. 2016, 17, 1790 10 of 16
Int. J. Mol. Sci. 2016, 17, 1790 10 of 16

Int. J. Mol. Sci. 2016, 17, 1790 10 of 16

Figure 1 H-NMR spectrum for the CS-102/PEO (80/20) mixture solubilized in 0.5 M acetic acid with
Figure 5.
5. 1H-NMR CS-102/PEO (80/20) mixture solubilized in 0.5 M acetic
Figure
added
added DD2OO5.(A)
1H-NMR spectrum for the CS-102/PEO (80/20) mixture solubilized in 0.5 M acetic acid with
(A) and CS-102
and CS-102 solubilized
solubilized in
in 0.5
0.5 M
M acetic
acetic acid
acid with
with added
added D O (B).
D22O (B).
added D2O (A) and CS-102 solubilized in 0.5 M acetic acid with added D2O (B).
After electrospinning,
After electrospinning, the the fibers
fiberswere
weremaintained
maintainedatatambient
ambientconditions
conditions forfor one
one week;
week; excess
excess of
After electrospinning, the fibers were maintained at ambient conditions for one week; excess
of acetic
acetic acid acid
and and water
water were were evaporated
evaporated to to stabilize
stabilize the the Then,
material. material.
the Then, the composition
composition was was
investigated
of acetic acid and water were evaporated to stabilize the material. Then, the composition was
investigated
after afterin
dissolution dissolution in D2O in acid (DCl)conditions (DCl) (Figure
Firstly,6).a6).Firstly,
Firstly, ashift
chemical shift was
investigated after D 2 O in acidinconditions
dissolution D2O in acid conditions(Figure 6).(Figure
(DCl) chemical a chemicalwas noted
shift was for
noted
all the for all
signalsthe signals
compared compared
with with
those those
obtained obtained
in aceticin acetic
acid acid
(+0.2 ppm(+0.2 in ppm
DCl) in DCl)
and
noted for all the signals compared with those obtained in acetic acid (+0.2 ppm in DCl) and inversion and inversion
inversion in the
in the
respectiverespective
in the respectivechemical
chemicalchemical shifts
shifts ofshiftsof
the –CH the –CH
3 of
of the of
acetyl
–CH 3 acetyl
groups
3 of acetyl
groups (at
(at 2.15
groups 2.15
(at ppm)
2.15 ppm)ppm)
and and and
residualresidual acetate
acetate
residual salt
acetate salt
from
salt
from
acetic acetic
acid
from acid (at
(at 2.18
acetic 2.18
acid ppm). ppm).
(at 2.18 ppm).

Figure
Figure 6. As
6. As spunspun fibers
fibers from
from thethe mixtureCS/PEO
mixture CS/PEO (80/20)
(80/20) dissolved
dissolvedininacidic D2D
acidic O2(A); Stabilized
O (A); andand
Stabilized
washed
Figure
washed Aschitosan
6.chitosan
spunfibersfibers
fibers from
from
from the
the
the mixtureCS-102/PEO
mixture
mixture CS-102/PEO
CS/PEO (80/20)80/20
80/20 dissolved
dissolved
dissolved in in
in acidic
acidic D22O
D
acidic O2(B).
D (A); Stabilized and
O (B).
washed chitosan fibers from the mixture CS-102/PEO 80/20 dissolved in acidic D2O (B).
From spectrum A, the amount of acetate groups was estimated and corresponded to a large
From
part
spectrum
of spectrum
the chitosan
A, the amount of acetate groups was estimated and corresponded to a large part
−NH counterions formed in presence of acetic acid. The proton of acetyl
From +A, the amount of acetate groups was estimated and corresponded to a large
of the chitosan − NH counterions formed in presence of
in acetic acid. The proton of acetylprocesses.
substituents
partsubstituents
of the chitosanalso gave
3
−NH the DA, which
counterions was formed
not modified the electrospinning
in presence of acetic acid. and The
washing
proton of acetyl
alsoThe
gave PEO content in the material was clearly determined at 3.86 ppm but overestimated due The
the DA, which was not modified in the electrospinning and washing processes. PEO
to the
substituents also gave the DA, which was not modified in the electrospinning and washing processes.
content in the
overlap withmaterial was clearly
one chitosan proton. determined at 3.86with
After treatment ppm K2but
CO3overestimated
and water, thedue NMR to spectrum
the overlapwaswith
The PEO content in the material was clearly determined at 3.86 ppm but overestimated due to the
onesimplified
chitosan proton.
as shownAfter treatment
in Figure 6B. Thewith K2for
signal COthe
3 and water,
acetate –CHthe NMRwas
3 groups spectrum was simplified
nearly suppressed: the as
overlap with one chitosan proton. After treatment with K2CO3 and water, the NMR spectrum was
shownratioinbetween
Figure 6B. the The
–CHsignal
3 signals forofthe acetate
acetate and–CH 3 groups
acetyl equaledwas 1.60nearly suppressed:
in the initial theand
fibers (A) ratio between
became
simplified as shown in Figure 6B.only
Theasignal for the acetate –CH 3 groups was nearly suppressed: the
the –CH3 signals of acetate and acetyl equaled 1.60 in the initial fibers (A) and became lowerof
lower than 0.1 (B). In addition, small amount of PEO remained: the ratio of proton integral than
ratio–CH
betweenfromthe
2– addition,
PEO –CH
and3 signals
that of acetate
of chitosan and acetyl1.22/1.05
equaledwhich1.60 in the initialinfibers (A) andofbecame
0.1 (B). In only a small amountH-2 equaled
of PEO remained: the ratio is of
negligible comparison
proton integral of –CH2 –thefrom
lowersamethan 0.1in(B).
ratio In addition,
Figure 6A givingonly a small amount of PEO remained: the ratio of proton integral of
15.59/1.22.
–CH2– from PEO and that of chitosan H-2 equaled 1.22/1.05 which is negligible in comparison of the
same ratio in Figure 6A giving 15.59/1.22.
Int. J. Mol. Sci. 2016, 17, 1790 11 of 16

PEO and that of chitosan H-2 equaled 1.22/1.05 which is negligible in comparison of the same ratio in
Figure 6A giving 15.59/1.22.
1 H-NMR is an interesting technique for analyzing the composition of the fibers regardless of the

processing conditions. It is demonstrated that the treatment proposed allows obtaining nearly pure
chitosan nanofibers in the example discussed in this work.

2.4.2. Composition and Degree of Swelling of Chitosan Nanofibers

Comparison of the dried weight before and after the post-treatments allowed determination of
the amount of PEO extracted from the nanofibers. These data were confirmed by the NMR results.
In general, the post-treatment proposed in this work allowed us to obtain pure chitosan nanofibers
which were perfectly stable after neutralization in ethanol-aqueous medium at pH > 6.5. In these
conditions at neutral pH (immersion into salt, buffer and water), the cohesion of membrane was
retained and they were easy to manipulate. Then, the degree of swelling was determined with respect
to the purified never-dried chitosan fibers. An example is given for the chitosan fibers obtained with
the ratio 80/20. The degree of swelling based on pure chitosan nanofibers was equal to 5.0 ± 0.2 g/g
in water. In PBS buffer, the degree of swelling was equal to 4.0 ± 0.2 g/g in relation with ionic
concentration of the buffer. In the same conditions, after K2 CO3 in ethanol-water treatment and water
extraction, the degree of swelling of fibers produced from 60/40 mixture was 6.7 ± 0.1 g/g, probably
due to larger accessibility of the never dried fibers after the extraction of 40% PEO. In addition, the
dried weight was lower than the initial one, corresponding to the chitosan left as insoluble under
the –NH2 form after extraction of water soluble PEO as mentioned previously. In fact, it is very
important to determine the degree of swelling on the basis of the final dried weight of pure chitosan.
After drying, the fibers were rehydrated in pure water to determine the water regaining capacity.
The results give small decrease of the original swelling, with values of 4.0 ± 0.2 and 2.42 ± 0.2 g/g for
CS/PEO (80/20) and CS/PEO (60/40), respectively.

3. Materials and Methods

3.1. Materials
Two different chitosan (CS) samples were used in this work. The first sample of chitosan was used
as a moderate molecular weight (MW ) and it is identified as CS-102. Its commercial name is Seacure
143 with MW = 102 kg/mol and degree of acetylation (DA) = 0.12. This was obtained from Pronova
Biopolymer (now Novamatrix, Sandvika, Norway). The second sample had a higher molecular weight;
it is identified as CS-500. Its name is Kitomer and it is produced by Marinard Lte, Montreal, QC,
Canada) with MW = 500 kg/mol and DA = 0.19. PEO with different molecular weights M (8 × 103 ,
5 × 103 , 1 × 103 , 3 × 102 , 1 × 102 kg/mol), acetic acid (≥99.7 %), ethanol and K2 CO3 were purchased
from Sigma-Aldrich (Munich, Germany). Deionized water (from Egis Eau, Grenoble, France) was used
as solvent to make up the solutions. PBS buffer was provided by Gibco Life Technologies (osmolality
280–315 mOsm/kg; ref 10010023, Gaithersburg, MD, USA) for study of chitosan membrane stability.
All reagents and polymers were used as received without further purification.

3.2. Sample Preparation

Chitosan (CS-102 and CS-500) solutions were prepared separately at 5% (w/v) and 1.4% (w/v) in
0.5 M acetic acid, respectively. These solutions were prepared at room temperature under magnetic
stirring (100 rpm) for four days to obtain homogeneous solutions. In the same manner, poly(ethylene
oxide) with different molecular weights were solubilized at 5% w/v in 0.5 M acetic acid under magnetic
stirring. Each chitosan (CS) samples were mixed with the solution of PEO with weight ratio (in %) of
90/10, 80/20 70/30 and 60/40. Similarly, the same concentrations of CS-102 and CS-500 were blended
with powdered PEO. Then, chitosan concentration remained equal to 5% and 1.4% (w/v), respectively
with weight ratios of chitosan to PEO of 90/10, 80/20 70/30 and 60/40. The weight ratios were
Int. J. Mol. Sci. 2016, 17, 1790 12 of 16

expressed as weight of chitosan or PEO by total polymer weight. All the solutions were kept under
constant shaking at 100 rpm for one day at room temperature to achieve complete mixing. This addition
of powder (PEO) in the solution of chitosan was selected to avoid the dilution of the final chitosan in
the polymer blend solution when mixing the two polymer solutions. These solutions were kept at rest
more than 30 min prior to electrospinning in order to remove air bubbles at room temperature. Ionic
conductivities of the blended CS/PEO solutions were measured using a conductivity meter (CRISON
CM 35, Barcelona, Spain) at room temperature.

3.3. Electrospinning
The prepared solutions were placed in a 5 mL plastic syringe fitted with a 21-gauge stainless
steel needle. The syringe pump delivered solutions at specified flow rate vertically (model: KDS
Legato 200, KD Scientific, Holliston, MA, USA), and electrospun with an applied voltage of 20 kV
between the electrodes using a homemade dual high voltage power supplier (±30 kV, iseq GMBH,
Radeberg, Germany). Then, the nanofibers were deposited on a rectangular grounded aluminum foil
collector kept at either 10, 11, 13 or 15 cm away from at the tip of the needle. The experiments were
carried out at room temperature in closed Plexiglas® box with relative humidity ranging between
50% and 60%. The produced nanofibers matrices were left in ambient conditions to evaporate excess
of acetic acid and water prior to further analyses.

3.4. Characterization of Nanofibers

3.4.1. Morphology of the Nanofibers Membranes

The SEM analyses of the samples were performed at CMTC-INP, Grenoble, France. The
morphology of electrospun nanofibers membranes, including the washed samples were observed
with a scanning electron microscope (Zeiss ultra 55 SEM FEG, Oberkochen, Germany) operated at
1 kV. The nanofibers samples were coated with 1 nm gold/palladium layer for 10 s prior to SEM
imaging. The average fiber and bead diameter (AFD) was calculated by randomly selected diameter of
100 nanofibers/nano-beads from each sample.

3.4.2. Determination of Swelling Capacity

Weighted initial nanofiber membrane cut in pieces were immersed in 1 M K2 CO3 solution in
water or ethanol/water (70/30) mixture to neutralize the chitosan. Further, nanofibers membranes
were washed four times daily for three days with deionized water until neutral pH was achieved
in order to obtain removal of the salt formed from chitosan solutions (potassium acetate), K2 CO3
excess and PEO. The following step was stabilization in water or PBS buffer (pH = 7.4) for 24 h.
Lastly, the membranes were dried at room temperature for further determination of the swelling
capacity. This dried weight decreases compared to the initial dried weight indicating PEO extraction
as confirmed by NMR experiments.
The swelling of the nanofibrous membranes was examined in terms of water loss between swollen
state in water at neutral pH and final dried weight at room temperature. The wet swollen samples
were weighed after blotting with tissue paper to remove excess surface water (Ww ). Accordingly, the
dried samples were also weighted repeatedly until the mass became constant (Wd ). The measurements
were carried out three times each. The average data were taken for the determination of swelling ratio
S using the following equation:
Ww − Wd
 
S (g/g) =
Wd
where Ww (g) is the weight of the swollen nanofibrous mat and Wd (g) is the weight of the samples
after drying at room temperature.
Rehydration of dried samples was tested using the same conditions after two days in water at
room temperature.
Int. J. Mol. Sci. 2016, 17, 1790 13 of 16

3.4.3. Stabilization of Chitosan Nanofibers

The membrane stabilization was controlled by neutralization and regeneration of chitosan under
the –NH2 form, which is insoluble in water over pH = 6.5. For this purpose, two different approaches
were followed. The produced nanofibers from the blend of PEO (powder) added into CS-102 solution at
5% (w/v) were washed in aqueous or ethanol-aqueous mixture at (70/30) in the presence of 1 M K2 CO3 .
After these first treatments, the membranes were repeatedly washed to remove water soluble PEO and
salt excess. At the same time, the morphological change exhibited by extraction of PEO and swelling
of the matrices was controlled for each sample. The effect of morphology and composition of the
fibers changes were studied after drying at room temperature for 24 h by SEM and NMR, respectively.
In addition, prolonged stability up to six months of CS/PEO nanofiber structure was investigated after
initial immersion and neutralization of the membranes in 70/30 ethanol-aqueous solutions. In addition,
the extraction of PEO was confirmed by difference between the dried weight of the initial membrane
and the dried weight left after K2 CO3 , water or buffer post-treatments. The resulting stable material
would then be pure chitosan nanofibers.

3.4.4. NMR Characterization of Nanofibers

The composition of the nanofibers in chitosan, PEO and acetic acid (or acetate) remaining in the
fibers were determined by 1 H-NMR at 80 ◦ C on a Bruker Avance III 400 spectrometer (Bruker, Billerica,
MA, USA). Selected samples were used to analyze the NMR spectrum and exemplify the change in the
nanofibers compositions. Nanofiber samples of 7 to 10 mg were dissolved in 1 mL D2 O in presence of
a stoichiometric amount of DCl. In a different test, 0.2 mL of chitosan solution and of CS/PEO mixed
solution (80/20) in 0.5 M acetic acid were added to 1 mL D2 O. Analysis of spectra allows determination
of the amount of acetic acid remaining in the dried samples and yield of PEO remaining in the samples
after washing in different conditions.

3.4.5. Rheological Behavior

Rheological characteristics were studied using a ARG2 rheometer from TA Instruments
(New Castle, DE, USA) with cone-plate geometry; the cone has a diameter of 50 mm, a 2◦ angle and a
60-µm gap. The temperature is controlled at 20 ◦ C by a Peltier plate. Steady-state flow experiments
were performed in the range of 0.05 to 10 s−1 . The dynamic oscillation measurements were carried out
for frequency sweeps between 0.5 to 10 Hz with 1% strain, corresponding to the linear viscoelastic
region (LVR).

4. Conclusions
In this study, production and morphology of chitosan nanofibers were investigated using different
molecular weights of chitosan, PEO and the blend of different CS/PEO ratios in 0.5 M acetic acid
solution. The electrospinning process parameters were optimized to achieve stable chitosan nanofibers
from the blends of CS/PEO. Particularly, the blend of chitosan with molecular weight of 102 kg/mol
and PEO of 1000 kg/mol were used to produce nanofibers for further exploring the structural and
chemical changes. The results have demonstrated that CS/PEO solutions containing 20 wt % and more
than 20 wt % of powdered PEO solubilized into the chitosan 5% (w/w) solution improved spinnability,
producing bead-free nanofibers. It was found that decreasing chitosan content in CS/PEO ratio from
80/20 to 60/40 leads to an increase in the nanofiber diameters from 85 to 147 nm, respectively. Image
analyses on the nanofiber matrix from blends of CS/PEO (60/40) and (80/20) demonstrate that, after
neutralization in 70% of ethanol aqueous solution, chitosan fiber morphology is stable in the aqueous
solution at pH > 6.5. Particularly, the obtained nanofibers membranes preserved morphological
structure after repeated water washings while the opposite was observed by neutralization in water at
room temperature. In addition, the swelling capacity of these produced nanofibers of CS/PEO (80/20)
and (60/40) matrix was around 5 and 6.7 (g/g), respectively. Degree of rehydration was slightly lower
Int. J. Mol. Sci. 2016, 17, 1790 14 of 16

than these swelling values of never dried fibers. This suggests that chitosan nanofibers structures
can be a good biomaterial candidate for biomedical, food, pharmaceutical and other applications
due the use of toxic free solvents. In addition, the NMR analysis revealed that a complete extraction
of PEO, acetic acid, potassium acetate and carbonate contents from nanofibers matrix was attained.
In general, we successfully prepared pure stable electrospun chitosan nanofibers without using strong
and/or toxic solvents. This optimized process can be used for further production of blend of CS/PEO
nanofibers, particularly for stable chitosan nanofibers in wide range of applications.

Acknowledgments: This work received the financial support from the Institut Carnot PolyNat through the
project RoCoCo. “Laboratoire Rhéologie et Procédés” (LRP) is a part of the LabEx Tec 21 (Investissements
d’Avenir-grant agreement n◦ ANR-11-LABX-0030). The authors gratefully acknowledge Rachel Martin from
the CMTC, Isabelle Jeacomine and Claire Boisset from Centre de Recherches sur les Macromolecules Végétales
(CNRS, Grenoble, France) for their expert help and technical assistance in FESEM microscopy, NMR and SEC
experiments, respectively.
Author Contributions: Solomon Mengistu Lemma, Frédéric Bossard and Marguerite Rinaudo conceived,
designed, performed the experiments and analyzed the data.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Haider, A.; Haider, S.; Kang, I.K. A comprehensive review summarizing the effect of electrospinning
parameters and potential applications of nanofibers in biomedical and biotechnology. Arab. J. Chem. 2015.
[CrossRef]
2. Chen, J.P.; Chang, G.Y.; Chen, J.K. Electrospun collagen/chitosan nanofibrous membrane as wound dressing.
Colloids Surf. A 2008, 313, 183–188. [CrossRef]
3. Chen, Z.G.; Wang, P.W.; Wei, B.; Mo, X.M.; Cui, F.Z. Electrospun collagen–chitosan nanofiber: A biomimetic
extracellular matrix for endothelial cell and smooth muscle cell. Acta Biomater. 2010, 6, 372–382. [CrossRef]
[PubMed]
4. Zhang, Y.; Reddy, V.J.; Wong, S.Y.; Li, X.; Su, B.; Ramakrishna, S.; Lim, C.T. Enhanced biomineralization in
osteoblasts on a novel electrospun biocomposite nanofibrous substrate of hydroxyapatite/collagen/chitosan.
Tissue Eng. Part A 2010, 16, 1949–1960. [CrossRef] [PubMed]
5. Chen, J.P.; Chen, S.H.; Lai, G.J. Preparation and characterization of biomimetic silk fibroin/chitosan
composite nanofibers by electrospinning for osteoblasts culture. Nanoscale Res Lett. 2012, 7, 1–11. [CrossRef]
[PubMed]
6. Zhang, Y.; Venugopal, J.R.; El-Turki, A.; Ramakrishna, S.; Su, B.; Lim, C.T. Electrospun biomimetic
nanocomposite nanofibers of hydroxyapatite/chitosan for bone tissue engineering. Biomaterials 2008, 29,
4314–4322. [CrossRef] [PubMed]
7. Jalaja, K.; Naskar, D.; Kundu, S.C.; James, N.R. Potential of electrospun core–shell structured gelatin-chitosan
nanofibers for biomedical applications. Carbohydr. Polym. 2016, 136, 1098–1107. [CrossRef] [PubMed]
8. Jayakumar, R.; Prabaharan, M.; Nair, S.V.; Tamura, H. Novel chitin and chitosan nanofibers in biomedical
applications. Biotechnol. Adv. 2010, 28, 142–150. [CrossRef] [PubMed]
9. Torres-Giner, S.; Ocio, M.J.; Lagaron, J.M. Development of active Antimicrobial fiber-based chitosan
polysaccharide nanostructures using electrospinning. Eng. Life Sci. 2008, 8, 303–314. [CrossRef]
10. Subramanian, A.; Vu, D.; Larsen, G.F.; Lin, H.Y. Preparation and evaluation of the electrospun chitosan/PEO
fibers for potential applications in cartilage tissue engineering. J. Biomater. Sci. Polym. Ed. 2005, 16, 861–873.
[CrossRef] [PubMed]
11. Rinaudo, M. Chitin and chitosan: Properties and applications. Prog. Polym. Sci. 2006, 31, 603–632. [CrossRef]
12. Younes, I.; Rinaudo, M. Chitin and chitosan preparation from marine sources. Structure, properties and
applications. Mar. Drugs 2015, 13, 1133–1174. [PubMed]
13. Desai, K.; Kit, K.; Li, J.; Zivanovic, S. Morphological and surface properties of electrospun chitosan nanofibers.
Biomacromolecules 2008, 9, 1000–1006. [CrossRef] [PubMed]
14. Haider, S.; Park, S.Y. Preparation of the electrospun chitosan nanofibers and their applications to the
adsorption of Cu(II) and Pb(II) ions from an aqueous solution. J. Membr. Sci. 2009, 328, 90–96. [CrossRef]
Int. J. Mol. Sci. 2016, 17, 1790 15 of 16

15. Mokhena, T.C.; Jacobs, V.; Luyt, A.S. A review on electrospun bio-based polymers for water treatment.
Express Polym. Lett. 2015, 9, 839–880. [CrossRef]
16. Geng, X.; Kwon, O.H.; Jang, J. Electrospinning of chitosan dissolved in concentrated acetic acid solution.
Biomaterials 2005, 26, 5427–5432. [CrossRef] [PubMed]
17. Pakravan, M.; Heuzey, M.C.; Ajji, A. A fundamental study of chitosan/PEO electrospinning. Polymer 2011,
52, 4813–4824. [CrossRef]
18. Rieger, K.A.; Birch, N.P.; Schiffman, J.D. Electrospinning chitosan/poly(ethylene oxide) solutions with
essential oils: Correlating solution rheology to nanofiber formation. Carbohydr. Polym. 2016, 139, 131–138.
[CrossRef] [PubMed]
19. Homayoni, H.; Ravandi, S.A.H.; Valizadeh, M. Electrospinning of chitosan nanofibers: Processing
optimization. Carbohydr. Polym. 2009, 77, 656–661. [CrossRef]
20. Ojha, S.S.; Stevens, D.R.; Hoffman, T.J.; Stano, K.; Klossner, R.; Scott, M.C.; Gorga, R.E. Fabrication
and characterization of electrospun chitosan nanofibers formed via templating with polyethylene oxide.
Biomacromolecules 2008, 9, 2523–2529. [CrossRef] [PubMed]
21. Sencadas, V.; Correia, D.M.; Areias, A.; Botelho, G.; Fonseca, A.M.; Neves, I.C.; Gómez-Ribelles, J.L.;
Mendez, S.L. Determination of the parameters affecting electrospun chitosan fiber size distribution and
morphology. Carbohydr. Polym. 2012, 87, 1295–1301. [CrossRef]
22. De Vrieze, S.; Westbroek, P.; van Camp, T.; van Langenhove, L. Electrospinning of chitosan nanofibrous
structureseasibility study. J. Mater. Sci. 2007, 42, 8029–8034. [CrossRef]
23. Ohkawa, K.; Cha, D.; Kim, H.; Nishida, A.; Yamamoto, H. Electrospinning of chitosan. Macromol. Rapid
Commun. 2004, 25, 1600–1605. [CrossRef]
24. Sun, K.; Li, Z.H. Preparations, properties and applications of chitosan based nanofibers fabricated by
electrospinning. Express Polym. Lett. 2011, 5, 342–361. [CrossRef]
25. Sarhan, W.A.; Azzazy, H.M.E. High concentration honey chitosan electrospun nanofibers: Biocompatibility
and antibacterial effects. Carbohydr. Polym. 2015, 122, 135–143.
26. Schiffman, J.D.; Schauer, C.L. One-step electrospinning of cross-linked chitosan fibers. Biomacromolecules
2007, 8, 2665–2667. [CrossRef] [PubMed]
27. Sangsanoh, P.; Supaphol, P. Stability improvement of electrospun chitosan nanofibrous membranes in neutral
or weak basic aqueous solutions. Biomacromolecules 2006, 7, 2710–2714. [CrossRef] [PubMed]
28. Min, B.M.; Lee, S.W.; Lim, J.N.; You, Y.; Lee, T.S.; Kang, P.H.; Park, W.H. Chitin and chitosan nanofibers:
Electrospinning of chitin and deacetylation of chitin nanofibers. Polymer 2004, 45, 7137–7142. [CrossRef]
29. Jia, Y.T.; Gong, J.; Gu, X.H.; Kim, H.Y.; Dong, J.; Shen, X.Y. Fabrication and characterization of poly(vinyl
alcohol)/chitosan blend nanofibers produced by electrospinning method. Carbohydr. Polym. 2007, 67,
403–409. [CrossRef]
30. Zhang, Y.; Huang, X.; Duan, B.; Wu, L.; Li, S.; Yuan, X. Preparation of electrospun chitosan/poly (vinyl
alcohol) membranes. Colloid Polym. Sci. 2007, 285, 855–863. [CrossRef]
31. Dhandayuthapani, B.; Krishnan, U.M.; Sethuraman, S. Fabrication and characterization of chitosan-gelatin
blend nanofibers for skin tissue engineering. J. Biomed. Mater. Res. Part B: Appl. Biomater. 2010, 94, 264–272.
32. Pezeshki-Modaress, M.; Zandi, M.; Mirzadeh, H. Fabrication of gelatin/chitosan nanofibrous scaffold:
Process optimization and empirical modeling. Polym. Int. 2015, 64, 571–580.
33. Chen, Z.; Mo, X.; He, C.; Wang, H. Intermolecular interactions in electrospun collagen–chitosan complex
nanofibers. Carbohydr. Polym. 2008, 72, 410–418.
34. Chen, Z.; Mo, X.; Qing, F. Electrospinning of collagen-chitosan complex. Mater. Lett. 2007, 61, 3490–3494.
35. Gomes, S.R.; Rodrigues, G.; Martins, G.G.; Roberto, M.A.; Mafra, M.; Henriques, C.M.R.; Silva, J.C. In vitro
and in vivo evaluation of electrospun nanofibers of PCL, chitosan and gelatin: A comparative study.
Mater. Sci. Eng. C 2015, 46, 348–358. [CrossRef] [PubMed]
36. Duan, B.; Dong, C.; Yuan, X.; Yao, K. Electrospinning of chitosan solutions in acetic acid with poly(ethylene
oxide). J. Biomater. Sci. Polym. Ed. 2004, 15, 797–811. [CrossRef] [PubMed]
37. Spasova, M.; Manolova, N.; Paneva, D.; Rashkov, I. Preparation of chitosan-containing nanofibres by
electrospinning of chitosan/poly(ethylene oxide) blend solutions. e-Polymers 2004, 4, 624–635. [CrossRef]
38. Bhattarai, N.; Edmondson, D.; Veiseh, O.; Matsen, F.A.; Zhang, M. Electrospun chitosan-based nanofibers
and their cellular compatibility. Biomaterials 2005, 26, 6176–6184. [CrossRef] [PubMed]
Int. J. Mol. Sci. 2016, 17, 1790 16 of 16

39. Min, L.L.; Yuan, Z.H.; Zhong, L.B.; Liu, Q.; Wu, R.X.; Zheng, Y.M. Preparation of chitosan based electrospun
nanofiber membrane and its adsorptive removal of arsenate from aqueous solution. Chem. Eng. J. 2015, 267,
132–141. [CrossRef]
40. Wei, H.; Zhang, F.; Zhang, D.; Liu, Y.; Leng, J. Shape-memory behaviors of electrospun chitosan/
poly(ethylene oxide) composite nanofibrous membranes. J. Appl. Polym. Sci. 2015, 132. [CrossRef]
41. Kohsari, I.; Shariatinia, Z.; Pourmortazavi, S.M. Antibacterial electrospun chitosan–polyethylene oxide
nanocomposite mats containing bioactive silver nanoparticles. Carbohydr. Polym. 2016, 140, 287–298.
[CrossRef] [PubMed]
42. Pakravan, M.; Heuzey, M.C.; Ajji, A. Core-shell structured PEO-chitosan nanofibers by coaxial electrospinning.
Biomacromolecules 2012, 13, 412–421. [CrossRef] [PubMed]
43. Herold, D.A.; Keil, K.; Bruns, D.E. Oxidation of polyethylene glycols by alcohol dehydrogenase.
Biochem. Pharmacol. 1989, 38, 73–76. [CrossRef]
44. Frohbergh, M.E.; Katsman, A.; Botta, G.P.; Lazarovici, P.; Schauer, C.L.; Wegst, U.G.; Lelkes, P.I. Electrospun
hydroxyapatite-containing chitosan nanofibers crosslinked with genipin for bone tissue engineering.
Biomaterials 2012, 33, 9167–9178. [CrossRef] [PubMed]
45. Charernsriwilaiwat, N.; Opanasopit, P.; Rojanarata, T.; Ngawhirunpat, T.; Supaphol, P. Preparation
and characterization of chitosan-hydroxybenzotriazole/polyvinyl alcohol blend nanofibers by the
electrospinning technique. Carbohydr. Polym. 2010, 81, 675–680. [CrossRef]
46. Elsabee, M.Z.; Naguib, H.F.; Morsi, R.E. Chitosan based nanofibers, review. Mater. Sci. Eng. C 2012, 32,
1711–1726. [CrossRef]
47. Rinaudo, M.; Pavlov, G.; Desbrieres, J. Influence of acetic acid concentration on the solubilization of chitosan.
Polymer 1999, 40, 7029–7032. [CrossRef]
48. Nikolova, A.; Manolova, N.; Rashkov, I. Rheological characteristics of aqueous of solutions mixtures of
chitosan and polyoxyethylene. Polym. Bull. 1998, 41, 115–121.
49. Rošic, R.; Pelipenko, J.; Kocbek, P.; Baumgartner, S.; Bešter-Rogač, M.; Kristl, J. The role of rheology of
polymer solutions in predicting nanofiber formation by electrospinning. Eur. Polym. J. 2012, 48, 1374–1384.
[CrossRef]
50. Sano, M.; Hosoya, O.; Taoka, S.; Seki, T.; Kawaguchi, T.; Sugibayashi, K.; Morimoto, Y. Relationship between
solubility of chitosan in alcoholic solution and its gelation. Chem. Pharm. Bull 1999, 47, 1044–1046. [CrossRef]
51. Gu, B.K.; Park, S.J.; Kim, M.S.; Kang, C.M.; Kim, J.I.; Kim, C.H. Fabrication of sonicated chitosan nanofiber
mat with enlarged porosity for use as hemostatic materials. Carbohydr. Polym. 2013, 97, 65–73. [CrossRef]
[PubMed]
52. Heux, L.; Brugnerotto, J.; Desbrieres, J.; Versali, M.F.; Rinaudo, M. Solid state NMR for determination of
degree of acetylation of chitin and chitosan. Biomacromolecules 2000, 1, 746–751. [CrossRef] [PubMed]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).