Stability Improvement of Electrospun Chitosan Nanofibrous
Membranes in Neutral or Weak Basic Aqueous Solutions
Pakakrong Sangsanoh and Pitt Supaphol*
Technological Center for Electrospun Fibers and the Petroleum and Petrochemical College, Chulalongkorn
University, Soi Chula 12, Phyathai Road, Pathumwan, Bangkok 10300, Thailand
Received March 24, 2006; Revised Manuscript Received August 20, 2006
Further utilization of chitosan nanofibrous membranes that are electrospun from chitosan solutions in trifluoroacetic
acid (TFA) with or without dichloromethane (DCM) as the modifying cosolvent is limited by the loss of the
fibrous structure as soon as the membranes are in contact with neutral or weak basic aqueous solutions due to
complete dissolution of the membranes. Dissolution occurs as a result of the high solubility in these aqueous
media of -NH3+CF3COO-salt residues that are formed when chitosan is dissolved in TFA. Traditional
neutralization with a NaOH aqueous solution only maintained partial fibrous structure. Much improvement in the
neutralization method was achieved with the saturated Na2CO3aqueous solution with an excess amount of Na2-
CO3(s) in the solution. We showed that electrospun chitosan nanofibrous membranes, after neutralization in the
Na2CO3aqueous solution, could maintain its fibrous structure even after continuous submersion in phosphate
buffer saline (pH ) 7.4) or distilled water for 12 weeks.
In recent years, much attention has been paid on the use of
high electrostatic potentials in fabricating ultrafine fibers from
materials of diverse origins with diameters in the submicrometer
down to nanometer range by a process known as electrospin-
ning.1This process involves the application of a strong
electrostatic field across a conductive capillary attaching to a
reservoir containing a polymer liquid and a screen collector.
Upon increasing the electrostatic field strength up to a critical
value, charges on the surface of a pendant drop destabilize its
shape from partially spherical into conical. Beyond a critical
value of the electrostatic field strength, a charged polymer jet
is ejected from the apex of the cone. The ejected charged jet
accelerates toward the collector by the electrostatic forces, during
which the jet elongates and either dries out or solidifies to finally
leave ultrafine fibers on the collector.
To mimick natural tissues in vitro, the electrospinning
technique has been heavily explored, due mainly to its potential
for fabricating highly porous fibrous membranes with the
diameters of the individual fibers being in the range close to
the fibrous collagen bundles of about 30-130 nm found in the
natural extracellular matrix (ECM).2Because of the great
expectations for utilizing electrospun fibers in biomedical
applications, a number of natural and synthetic biodegradable
polymers have been electrospun: they are, for examples, native
collagens from calfskin and human placenta,2bovine fibrinogen,3
Bombyx mori and Samia cynthia ricini silk fibroins,4dextran,
methacrylated dextran, and dextran/poly(D,L-lactide-co-gly-
colide) (PLGA) hybrids,5poly(ester urethane)urea (PEUU)/
bovine collagen type I hybrids,6PLGA/chitin and poly(glycolic
acid) (PGA)/chitin hybrids,7and hyaluronic acid.8
Chitosan or poly(N-acetyl-D-glucosamine-co-D-glucosamine)
is a partially N-deacetylated derivative of chitin or poly(N-acetyl-
D-glucosamine), one of the most abundant polysaccharides. Even
though chitin is structurally similar to glycosaminoglycans
(GAGs), such as condroitin sulfate and hyaluronic acid in the
ECM,9its utilization is limited by its poor solubility and its
physical properties that are rigid and brittle. Chitosan has been
explored as a suitable functional material for biomedical
utilization, mainly due to its biocompatibility, biodegradability,
and nontoxicity.10Electrospinning of chitosan has proven to be
difficult; therefore, electrospinning of chitosan fibers has been
in blends with another polymer, such as poly(ethylene oxide)
(PEO) in an aqueous solution of acetic acid11,12or water,13silk
fibroin in an aqueous solution of formic acid,14and poly(vinyl
alcohol) (PVA) in an aqueous solution of formic acid15or acetic
Nevertheless, successful fabrication of pure chitosan nanofi-
bers has been reported from the electrospinning of chitosan
solutions in trifluoroacetic acid (TFA) or a cosolvent system of
TFA and dichloromethane (DCM),15deacetylation of chitin
nanofibers obtained from the electrospinning of chitin solutions
in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP),17and the electro-
spinning of chitosan solutions in 90% aqueous acetic acid
solution.18To further explore the use of electrospun chitosan
fibrous membranes as tissue scaffolds, we unsuccessfully tried
to electrospin the chitosan solutions in concentrated acetic acid
solutions.18Since the molar mass of chitosan is a critical
parameter determining whether uniform chitosan fibers can be
obtained,18it is postulated that our failure was a result of the
unsuitable molar mass of the chitosan sample used. Subse-
* To whom correspondence should be addressed. E-mail: pitt.s@
Biomacromolecules 2006, 7, 2710-2714
10.1021/bm060286l CCC: $33.50© 2006 American Chemical Society
Published on Web 09/22/2006
quently, we had a success with a chitosan solution in both TFA
and TFA/DCM solvent systems,15but the resulting chitosan
membranes lost their fibrous structure as soon as they were in
contact with phosphate buffer saline (PBS; pH ) 7.4) or even
70% ethanol during sterilization. The loss in the fibrous structure
of the chitosan membranes is thought to be a result of the
dissolution of the chitosan trifluoroacetate salts that are formed
when chitosan is dissolved in TFA.19
Since we found that the fabrication of pure chitosan
nanofibrous membranes from chitosan solutions in TFA or TFA/
DCM was relatively easy, a neutralization procedure is needed
to further realize the actual usefulness of the membranes in areas
that require a contact of the membranes with neutral or weak
basic aqueous media. In the present contribution, we report an
alternative method for neutralizing the electrospun chitosan
nanofibrous membranes that were fabricated from a chitosan
solution in TFA or TFA/DCM. The effect of the neutralization
on morphology and stability in PBS of the as-spun chitosan
membranes was also investigated.
2. Experimental Section
2.1. Materials. Chitosan powder [degree of deacethylation (DD)
) 95%], trifluoroacetic acid (TFA, CF3COOH; ∼98% purity), and
dichloromethane (DCM) were purchased from Sigma-Aldrich (USA).
NaOH and Na2CO3, used as neutralizing agents, were purchased from
Sigma-Aldrich (USA). Phosphate buffer saline (PBS; pH ) 7.4), used
as the medium for weight loss and swelling assessments, was purchased
from Sigma-Aldrich (USA). Both the weight-average and the number-
average molar masses of chitosan were determined using a Waters 600E
size exclusion chromatograph (medium ) 0.5 M acetate buffer,
column: Ultrahydrogel linear, detector: refractive index, temperature
) 30 °C, and software: PL LogiCal) to be about 570 000 and 70 000
2.2. Fabrication of Electrospun Chitosan Nanofibrous Mem-
branes. Electrospun chitosan nanofibrous membranes were prepared
in a manner similar to that reported previously by Ohkawa et al.15
Briefly, 7% w/v chitosan solution was prepared by dissolving a
measured amount of chitosan powder in a mixture of TFA and DCM
(70:30 v/v). The as-prepared chitosan solution was continuously stirred
for 12 h at room temperature and later fed into a 5-mL glass syringe
fitted with a gauge 20 stainless steel needle (OD ) 0.91 mm) used as
the nozzle. Both the syringe and the needle were tilted 45° from a
vertical baseline. The as-spun chitosan nanofibers were collected on
an aluminum sheet wrapped around a homemade rotating cylinder
(width and diameter ≈ 15 cm), placed at a fixed distance of 20 cm
from the needle tip. The needle was connected to the emitting electrode
of positive polarity of a Gamma High-Voltage Research ES30P-5W
power supply. Both the electrical potential and the collection time were
fixed at 25 kV and 24 h and the solution feed was driven mainly by
the gravity and the electrostatic forces generated during spinning. The
resulting chitosan nanofibrous membranes were dried in vacuo at room
temperature prior to further investigation.
2.3. Neutralization Treatments. Neutralization of the as-spun
chitosan nanofibrous membranes was carried out by immersing the
membranes in either 5 M NaOH or 5 M Na2CO3aqueous solution for
3 h at ambient condition. “5 M Na2CO3aqueous solution” was used in
a loose term: since the solubility limit of the Na2CO3salt in water is
about 33%, the as-prepared 5 M of the solution then comprised a
saturated Na2CO3aqueous solution and an excess amount of Na2CO3-
(s). After the immersion, the membranes were repeatedly washed with
distilled water until neutral pH was obtained, dried at ambient condition
for 1 d, and further dried in an oven at 40 °C overnight prior to further
2.4. Characterization of Pre- and Post-Neutralized Chitosan
Nanofibrous Membranes. To evaluate the effectiveness of the
neutralization treatments, the morphology of the as-spun chitosan
nanofibrous membranes was investigated by a JEOL JSM-5200
scanning electron microscope (SEM). Five samples for each neutraliza-
tion treatment were coated with gold by a JEOL JFC-1100E sputtering
device for 3 min prior to SEM observation. A Nicolet Nexus 671 Fourier
transformed infrared spectroscope (FT-IR) was used to verify the
chemical structure of both the pre- and the post-neutralized membranes.
X-ray diffraction (XRD) was also used to observe the packing nature
of chitosan both before and after the neutralization treatments. X-ray
diffraction was carried out on a Rigaku Rint2000 X-ray diffractometer
over the 2θ range of 5-90° at a scanning speed of 5°/min.
Additionally, the physical integrity in terms of weight loss and
swelling of the post-neutralized chitosan nanofibrous membranes after
submersion in PBS at various time intervals [i.e., up to 12 weeks (for
the weight loss measurement) or 80 min (for the degree of swelling
measurement)] was investigated. The results were also compared with
those obtained from chitosan films having a similar thickness. The
chitosan films were cast from a chitosan solution in 1% acetic acid
aqueous solution in a glass Petri dish, dried at ambient condition for 2
d, neutralized with 5 M NaOH aqueous solution for 3 h, dried at ambient
condition for 1 d, and finally dried in an oven at 40 °C overnight prior
to further characterization. Both the nanofibrous membrane and the
film samples were submerged in PBS for 24 h at ambient condition
prior to official timing of both the weight loss and the swelling
The weight loss (%) of each sample (circular disk of about 2 cm in
diameter) was calculated according to the following equation:
where Wdidenotes the initial weight of the sample in its dry state prior
to submersion in PBS and Wdtdenotes the weight of the sample in its
dry state after submersion in PBS for an arbitrary time interval. The
swelling behavior of each sample (circular disk of about 2 cm in
diameter) was assessed by gravimetric method. Each sample, after
submersion in PBS for an arbitrary time interval, was taken out and
placed between two pieces of tissue paper. A flat metal sheet (300 g)
was placed on top of the sample to remove excess PBS. The degree of
swelling (%) of each sample was then calculated according to the
where Wst denotes the weight of the sample in its wet state after
submersion in PBS for an arbitrary time interval.
3. Results and Discussion
Electrospinning of 7% w/v chitosan solution in 70:30 v/v
trifluoroacetic acid/dichloromethane (TFA/DCM) was relatively
easy. Ohkawa et al.15pointed out that the successful electro-
spinning of the chitosan solution in TFA was likely a result of
the formation of salts between TFA and amino groups along
the chitosan chain,19causing the rigid interaction between
chitosan molecules to decrease, thus improving the electrospin-
nability of the solution. A selected SEM image of the as-spun
chitosan fibrous membranes is shown in Figure 1a. Clearly,
fibers with smooth and bead-free structure were obtained.
Statistical analysis of the measured diameters showed the values
of 130 ( 10 nm. After consecutive spinning for 24 h, the
thickness of the obtained membranes was 20 ( 3 µm. In an
attempt to assess the biological compatibility with mammalian
cells, these chitosan nanofibrous membranes failed to maintain
their fibrous structure due to the complete dissolution of the
weight loss (%) )(Wdi- Wdt)
× 100 (1)
degree of swelling (%) )(Wst- Wdt)
CommunicationsBiomacromolecules, Vol. 7, No. 10, 2006
fibers when they came into contact with PBS or even the
sterilized 70% ethanol solution (see additional experiment in
the Supporting Information).
Upon the dissolution of chitosan in TFA, the formation of
salts between TFA molecules and the amino groups of chitosan
is thought to occur in two sequential steps: (1) protonation of
the amino (-NH2) groups along the chitosan chains and (2)
ionic interaction between the protonated amino (-NH3+) groups
and trifluoroacetate anions19and these salts are readily soluble
in an aqueous medium. To fully utilize the as-spun chitosan
nanofibrous membranes in applications that require a contact
with an aqueous medium, it is necessary to explore a possibility
to overcome the dissolution problem. The most logical way is
through deprotonation of the amino groups through a treatment
in an alkaline solution.19
Figure 1b shows a selected SEM image of a chitosan
nanofibrous membrane that was treated in 5 M NaOH aqueous
solution for 3 h. Evidently, even after the chitosan nanofibrous
membrane was neutralized with the NaOH aqueous solution,
its initial fibrous structure (see Figure 1a) was lost (see additional
experiment in the Supporting Information). Scheme 1 delineates
the reactions that might occur during the neutralization of
-NH3+CF3COO-salt residues along the chitosan chains with
NaOH(aq). As soon as the chitosan nanofibers were in contact
with NaOH(aq), the salt residues dissolved, leaving -NH3+
groups on the chitosan chains. Some of these groups would be
deprotonized with -OH ions to leave -NH2 groups on the
chitosan chains, whereas others would become hydrated. Based
on this postulated mechanism, chitosan nanofibers would
become either partially or completely dissolved after neutraliza-
tion with a NaOH solution, depending mainly on %DD and
molar mass of chitosan, concentration of the NaOH aqueous
solution, and diameters of the as-spun chitosan nanofibers.
Further improvement in the neutralization treatment of the
as-spun chitosan nanofibrous membranes was made by sub-
merging the membranes in 5 M Na2CO3aqueous solution for 3
h. Figure 1c shows a selected SEM image of such a membrane.
Apparently, the nanofibrous structure of the membrane was
intact after such a treatment (see additional experiment in the
Supporting Information). The reactions that might occur during
the neutralization of -NH3+CF3COO-salt residues with
Na2CO3(aq) on the chitosan chains are also described in Scheme
1. Similarly, as soon as the chitosan nanofibers were in contact
with Na2CO3(aq), the salt residues dissolved to leave -NH3+
groups on the chitosan chains. Deprotonation of the -NH3+
groups would occur very rapidly such that the detached proton
would react with CO32-ions to become HCO3-ions. In
addition, the detached proton can further react with HCO3-ions
to finally obtain carbonic acid, H2CO3. Due to the excess amount
of Na2CO3(s) in the as-prepared solution, neutralization of the
salt residues can continue until no residues are available (i.e.,
complete neutralization). Indeed, the post-neutralized chitosan
nanofibrous membranes still maintained their nanofibrous
structure even after being submerged in PBS for 12 weeks (see
Figure 1d). Though not shown, a similar result was also
observed on the post-neutralized membranes that were sub-
merged in distilled water for the same period.
Figure 2 shows the FT-IR spectra of the as-spun chitosan
nanofibrous membranes before and after neutralization with the
Na2CO3aqueous solution in comparison with that of the as-
received chitosan powder. The characteristic absorption peaks
of the pre-neutralized chitosan membrane were observed at 1675
and 1530 cm-1, corresponding to the stretching of the protonated
amino (-NH3+) groups. Evidently, the presence of the large
absorption peak at 1675 cm-1and the three absorption peaks
around 840-720 cm-1are indicative of the presence of
trifluoroacetic acid in chitosan nanofibers as amine salts.19On
the other hand, the post-neutralized chitosan nanofibers and the
as-received chitosan powder exhibited strong absorption peaks
at 3300 and 3400 cm-1, corresponding to the stretching of the
amino (-NH2) groups.17Evidently, FT-IR results confirmed the
regeneration of the amino groups after the treatment. An
improvement in the packing ability of chitosan molecules in
Figure 1. Selected SEM images of (a) pre-neutralized as-spun
chitosan nanofibrous membrane from 7% chitosan solution in 70:30
v/v TFA/DCM, chitosan nanofibrous membrane after neutralization
with (b) 5 M NaOH or (c) 5 M Na2CO3 aqueous solution, and (d)
chitosan nanofibrous membrane in panel c after submersion in PBS
for 12 weeks.
Biomacromolecules, Vol. 7, No. 10, 2006Communications
the fibers after neutralization was also supported by XRD
analysis (see the Supporting Information), in which the regen-
eration of the amino groups after neutralization that resulted in
the reestablishment of intermolecular hydrogen interaction
between chitosan molecules improved the molecular packing.
The effect of neutralization on both the weight loss and the
swelling behavior of the as-spun chitosan nanofibrous mem-
branes was also investigated (see Figures 3 and 4, respectively).
Without the treatment with the Na2CO3aqueous solution, the
as-spun membranes dissolved completely in PBS and distilled
water almost instantaneously. After the treatment, the loss in
the weight of the fibrous membrane samples submerged in PBS
increased very rapidly during the first three weeks and increased
gradually until it leveled off after about 6 weeks. For compari-
son, the loss in the weight of the solution-cast film samples
(thickness ) 21 ( 2 µm) also increased very rapidly during
the first three weeks in submersion, but, after three weeks, no
significant change was observed. After 12 weeks, the loss in
the weight of the film samples was about half of that of the
fibrous samples (i.e., 9 versus 16%, respectively), a result of
the very much greater surface area of the fibrous membranes
in comparison with that of the films. In analogy to the weight
loss, the degree of swelling for both the fibrous membrane and
the film samples increased initially with submersion period, but
became unchanged after a certain submersion period. Obviously,
the swelling of the fibrous membranes increased very rapidly
during the first 20 min and started to level off only after 30
min, with the highest degree of swelling being about 100%.
On the other hand, the swelling of the films increased
monotonically, but less rapidly in comparison with that of the
Scheme 1. Possible Chemical Reactions that Occur during Neutralization of As-Spun Chitosan Nanofibrous Membranes with (a) 5 M NaOH
or (b) 5 M Na2CO3Aqueous Solution
Figure 2. FT-IR spectra of as-spun chitosan nanofibrous membranes
before and after neutralization with 5 M Na2CO3 aqueous solution
for 3 h in comparison with that of as-received chitosan powder.
CommunicationsBiomacromolecules, Vol. 7, No. 10, 2006
fibrous membranes, during the first 50 min and started to level Download full-text
off after about 60 min, with the highest degree of swelling being
As an actual example, Figure 5 shows a selected SEM image
of a Schwann cell (RT4-D6P2T cell line) that was allowed to
attach on a post-neutralized chitosan nanofibrous membrane for
1 h. Clearly, the fibrous structure of the membrane was
maintained after sterilization with 70% ethanol, multiple washing
with PBS, and submersion in the culture medium. This confirms
the necessity of the neutralization to further explore the
usefulness of the electrospun chitosan nanofibrous membranes
in biomedical and other applications.
Acknowledgment. The authors acknowledge partial support
received from (1) the National Research Council of Thailand
(NRCT), (2) Chulalongkorn University (through invention and
research grants from the Ratchadapesek Somphot Endowment
Fund), (3) the Petroleum and Petrochemical Technology Con-
sortium [through a Thai governmental loan from Asian Devel-
opment Bank (ADB)], and (4) the Petroleum and Petrochemical
College (PPC), Chulalongkorn University. Also, Dr. Poonlarp
Cheepsunthorn of the Faculty of Medicine, Chulalongkorn
University is acknowledged for his expert guidance on the
culture of Schwann cells.
Supporting Information Available. Additional experiment
including results in Figure I and Table I. This material is
available free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) (a) Li, D.; Xia, Y. AdV. Mater. 2004, 16, 1151. (b) Frenot, A.;
Chronakis, I. S. Curr. Opin. Colloid Interface Sci. 2003, 8, 64. (c)
Huang, Z. M.; Zhang, Y. Z.; Kotaki, M.; Ramakrishna, S. Composite
Sci. Technol. 2003, 63, 2223.
(2) Matthews, J. A.; Wnek, G. E.; Simpson, D. G.; Bowlin, G. L.
Biomacromolecules 2002, 3, 232.
(3) Wnek, G. E.; Carr, M. E.; Simpson, D. G.; Bowlin, G. L. Nano Lett.
2003, 3, 213.
(4) Ohgo, K.; Zhao, C.; Kobayashi, M.; Asakura, T. Polymer 2003, 44,
(5) Jiang, H.; Fang, D.; Hsiao, B. S.; Chu, B.; Chen, W. Biomacromol-
ecules 2004, 5, 326.
(6) Stankus, J. J.; Guan, J.; Wagner, W. R. J. Biomed. Mater. Res. 2004,
(7) (a) Min, B. M.; You, Y.; Kim, J. M.; Lee, S. J.; Park, W. H.
Carbohydr. Polym. 2004, 57, 285. (b) Park, K. E.; Kang, H. K.; Lee,
S. J.; Min, B. M.; Park, W. H. Biomacromolecules 2006, 7, 635.
(8) Um, I. C.; Fang, D.; Hsiao, B. S.; Okamoto, A.; Chu, B. Biomac-
romolecules 2004, 5, 1428.
(9) Muzzareli, R. A. A. Encyclopedia of Polymer Science and Engineer-
ing, 3rd ed.; Wiley: New York, 1985; pp 430-440.
(10) Ravi Kumar, M. React. Funct. Polym. 2000, 46, 1.
(11) Duan, B.; Dong, C.; Yuan, X.; Yao, K. J. Biomater. Sci.-Polym.
Ed. 2004, 15, 797.
(12) Bhattarai, N.; Edmondson, D.; Veiseh, O.; Matsen, F. A.; Zhang,
M. Biomaterials 2005, 26, 6176.
(13) Spasova, M.; Manolova, N.; Paneva, D.; Rashkov, L. e-Polymers
2004, No. 056.
(14) Park, W. H.; Jeong, L.; Yoo, D. I.; Hudson, S. Polymer 2004, 45,
(15) Ohkawa, K.; Cha, D.; Kim, H.; Nishida, A.; Yamamoto, H.
Macromol. Rapid Commun. 2004, 25, 1600.
(16) Li, L.; Hsieh, Y. L. Carbohydr. Res. 2006, 341, 374.
(17) Min, B. M.; Lee, S. W.; Lim, J. N.; You, Y.; Lee, T. S.; Kang, P.
H.; Park, W. H. Polymer 2004, 45, 7137.
(18) Geng, X.; Kwon, O. H.; Jang, J. Biomaterials 2005, 26, 5427.
(19) Hasegawa, M.; Isogai, A.; Onabe, F.; Usuda, M. J. Appl. Polym.
Sci. 1992, 45, 1857.
Figure 3. Weight loss in PBS as a function of submersion time for
chitosan nanofibrous membranes after neutralization with 5 M Na2-
CO3 aqueous solution for 3 h and solution-cast chitosan films
(neutralized with 5 M NaOH aqueous solution for 3 h).
Figure 4. Degree of swelling in PBS as a function of submersion
time for chitosan nanofibrous membranes after neutralization with 5
M Na2CO3aqueous solution for 3 h and solution-cast chitosan films
(neutralized with 5 M NaOH aqueous solution for 3 h).
Figure 5. Selected SEM image of Schwann cell (RT4-D6P2T cell
line) that was allowed to attach for 1 h on chitosan nanofibrous
membrane after neutralization with 5 M Na2CO3 aqueous solution
for 3 h.
Biomacromolecules, Vol. 7, No. 10, 2006Communications