International Journal of Biological Macromolecules 75 (2015) 239247

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

journal homepage: www.elsevier.com/locate/ijbiomac

Curcumin/cellulose micro crystals/chitosan lms: Water absorption

behavior and in vitro cytotoxicity
S.K. Bajpai a, , Navin Chand b , Sonam Ahuja a , M.K. Roy c
a
Polymer Research Laboratory, Department of Chemistry, Govt. Model Science College, Jabalpur, MP 482001, India
b
Advanced Materials Processes and Research Institute, Polymer Composite Group, Bhopal, MP, India
c
IIITDM, Jabalpur India

a r t i c l e i n f o a b s t r a c t

Article history: A new technique, called vapor induced phase inversion (VIPI), has been employed to fabricate cellulose
Received 15 November 2014 micro crystals (CMC)-loaded chitosan (Ch) lms. The method involves immediate exposure of CMC-
Received in revised form 16 January 2015 dispersed chitosan solution to NH3 gas. The lms were characterized by Fourier Transform Infrared
Accepted 26 January 2015
Spectroscopy (FTIR), Scanning Electron Microscopy (SEM) and Atomic Force Microscopy (AFM) analysis.
Available online 30 January 2015
The swelling ratio (SR) of lms showed negative dependence on the cellulose content in the lms. The
dynamic water uptake data were interpreted by various kinetic models. Finally, the release of curcumin
Keywords:
from the lms was investigated. The CMC-loaded chitosan lm showed slower release as compared to
Phase inversion
Chitosan
the plain chitosan lm, suggesting that cellulose micro crystals acted as diffusion barrier. The lms were
Cellulose non-cytotoxic, non-thrombogenic and non-hemolytic.
Moisture permeation 2015 Elsevier B.V. All rights reserved.
Wound dressings

1. Introduction
Wound healing is the process in which damaged epidermal
tissues are replaced/regenerated by living issues [1]. A wound
dressing consists of a natural or synthetic polymer matrix, and
it absorb exudates with simultaneous release of the entrapped
bioactive agent at a pre-determined rate. An ideal wound dressing
prevents the wound from microbial contamination and enhances
the wound healing process. An ideal wound dressing should be
able to permeate the gases and keep the wound environment
moist. In addition, it should be biocompatible and preferable
biodegradable [2]. In the last decade, there has been sincere concern
about saving the aquatic environment from synthetic polymers,
and therefore, the focus of research has now switched over to
using naturally occurring polysaccharides for biomedical applica-
tions [3]. In recent past, bio-polymers such as cellulose, chitin,
chitosan, sodium alginate, collagen, gelatin etc. have been used
frequently in wound dressings and other biomedical applications
[4]. The major advantages of natural polymers include their good
cytocompatibility, biocompatibility, and above all, their biodegrad-
able nature which eliminates chances of post-operative surgery
for removal of implanted polymeric device. Gelatin, chitosan
Corresponding author. Tel.: +91 09993220651.
E-mail address: sunil.mnlbpi@gmail.com (S.K. Bajpai).
[5], collagen [6] etc. are among popular dressing materials used
for quicker wound healing. Chitosan [-(14)-2-amino-2-deoxy-d-
glucose], a de-acetylated form of chitin is a naturally occurring
linear biodegradable polysaccharide and it is made up of N-acetyl-
d-glucosamine and d-glucosamine. It is nontoxic, biocompatible
[7], biodegradable polymer [8]. It is used in drug delivery, cell
delivery systems, orthopedics, wound healing [9], ophthalmology,
and bone healing [10]. It has ability to form complexes with DNA
and poly anionic polymers [11], and can open intercellular tight
junctions and it is totally biocompatible. It exhibits antimicrobial
activity against bacteria [12], fungi, and yeast. A chitosan lm with
controllable water absorption and drug release properties can be
obtained using cross linking agents such as glutaraldehyde [13].
However, a number of reports are available which claim its severe
toxicity [14]. Therefore, in place of GTA, other cross linkers such as
genipin [15], suberoyl chloride [16] epichlorohydrin [17] etc. have
also been used, but they are also reported to be toxic for human [18].
With the aim to fabricate un-cross linked chitosan lm, but with
controllable physical and chemical properties, we have dispersed
cellulose micro crystals uniformly into chitosan lm using a novel
VIPI approach. In this method, CMC dispersed chitosan solution is
exposed to NH3 gas. This causes a sudden change in pH of the dope
and chitosan is precipitated as porous lm, with homogeneously
distributed cellulose micro crystals. These cellulose crystals act
as diffusion barrier to control the swelling, moisture permeation
and drug release properties of the resulting lms. The CMC/Ch

http://dx.doi.org/10.1016/j.ijbiomac.2015.01.038
0141-8130/ 2015 Elsevier B.V. All rights reserved.
240 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247
lms have been loaded with natural bioactive Ingredient curcumin,
which is a hydrophobic polyphenolic compound derived from the
rhizome of the perennial herb Curcuma longa [19]. It is reported
to possess a number of biological activities like wound healing,
antibacterial, antioxidant, anti-inammatory, anticancer proper-
ties etc. [20].
2. Materials and method
2.1. Materials
Cellulose micro crystals (CMC), chitosan (Ch) (degree of de-
acetylation 98%, molecular weight 1.23 105 g/mol), glacial acetic
acid and liquid ammonia were obtained from E. Merck, Mumbai,
India. Other chemicals were purchased from SRL, Pune, India and
were of analytical grade. The distilled water was used throughout
the investigations.
2.2. Fabrication of CMC/Ch lm by VIPI approach
An in-lab built apparatus, employed in VIPI approach, is shown
in Fig. 1.
A 250 mL conical ask, containing liquid NH3 was tted with an
airtight wooden cork having a hole of diameter 1 cm. A glass tube,
bent through right angle at one place was placed in the cork hole
and its other end was jointed below the mouth of the glass jar and
made airtight. The conical ask was warmed gently at 40 C. The
NH3 gas passed into the glass jar and created NH3 environment. In
order to prepare CMC/Ch composite lm, 0.5 g of Ch was dissolved
in 25 mL of 2% (w/v) solution of acetic acid and to this different
amount of cellulose micro crystals (CMC), (i.e. 10, 20 and 30 wt% of
the chitosan content) were added. The suspension was agitated on
a magnetic stirrer (Remi, India) for 30 min to ensure uniform mix-
ing, poured into a Petri plates and immediately placed in the gas
jar. The Petri plates were taken out after 5 min and put in an electric
oven at 60 C for complete evaporation of solvent. In all, four lms
were fabricated, whose designations were Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20), and Ch/CMC(30) respectively. The number in paren-
thesis denotes the wt% of cellulose micro crystals (CMC) relative to
the chitosan in the feed mixture.
2.3. Extraction of curcumin (CC) from turmeric
The extraction of curcumin from turmeric, using polar organic
solvents such as alcohols, acetone, ethyl acetate, etc. has been
reported by a number of workers [21]. However, according to a
report by Revathy et al. [22], out of a number of extraction solvents
like acetone, chloroform, hexane, methanol, ethyl acetate, the sol-
vent acetone yielded maximum curcumin recovery of almost 22.8%.
Therefore, we also decided to extract curcumin from turmeric using
acetone following the method proposed by Saraswathi et al. [23].
In brief, 20 g of ne turmeric powder was suspended in 150 mL of
acetone under moderate stirring for 72 h at 37 C. The mixture was
ltered, the ltrate was poured into a Petri plate and the solvent
was evaporated under vacuum to obtain semi-dry oily mass. The
oily mass was weighed accurately (2.99 g) and dissolved in 50 mL
of DMSO to give a reddish-brown curcumin solution. The curcumin
content in the above extract was calculated using the following
expression
(amount of oily mass)
Curcumin content per mL = =
volume of DMSO
2.99
= g/mL
50
= 0.0598g/mL
In order to prepare a lm, 3.0 mL of extract was used, which
contained 0.18 g of curcumin.
2.4. Preparation of curcumin (CC) loaded samples
In order to prepare CC loaded lms, 3 mL of above extract
(containing 0.18 g of CC) was added into CMC/Ch dispersion
under constant stirring, followed by transfer into Petri plates.
The Petri plates were exposed to NH3 gas for lm formation as
described above. The samples were designated as Ch/CMC(0)0.18
and Ch/CMC(20)0.18 where the samples carry their usual meaning
and the sub-script is the amount of CC in g loaded in the lm.
2.5. Characterization of lms
The Fourier Transform Infrared (FTIR) spectra were recorded
with an FTIR spectrophotometer (Shimadzu, 8400, Japan) using
KBr. The surface morphology of all the lms was determined by
Scanning Electron Microscopy (SEM) and Atomic Force Microscopy
(AFM) at Indian Institute of Technology and Design of Materials,
Jabalpur (India).
2.6. Swelling studies of lms
The swelling behavior of lms was studied in the pseudo extra-
cellular uid (PEF), with the following composition: 1 L of the
solution contained 2.2 g of KCl, 6.8 g of NaCl, 25 g of sodium bi-
carbonate and 3.5 g of sodium di-hydrogen phosphate. The pH
of this solution was 7.36. A pre-weighed lm sample was placed
in 100 mL of PEF at 37 C and it was taken out at different time
intervals, wiped supercially with tissue paper to remove extra
surface water, weighed accurately in an electronic balance (Den-
ber, Germany), and then placed again in water. The swelling ratio
(SR) was using the following expression:
(Mt M0 )
SR = g/g (2)
M0
where M0 and Mt are the initial mass and mass at different time
intervals respectively. To determine equilibrium swelling ratio
(ESR), Mt was replaced by Me which is the weight of the swollen
lm at equilibrium.
2.7. Dynamics of moisture uptake
The kinetics of moisture uptake was performed using the
method used elsewhere [24]. Saturated solution of KNO3 was put
in a plastic jar and a rectangular block of Stainless steel, with its
top above the level of the solution, was put in the middle of the jar.
Now an aluminum crucible was placed on the top of the steel block.
A pre-weighed piece of completely dry lm was placed on the cru-
cible, lid of the jar was closed tightly and the jar was placed in an
incubator, maintained at 37 C. The lm was taken out at regular
time intervals, weighed accurately by an electronic balance (Den-
ber, Germany) and then put back in the jar. The mass measurements
were continued till the attainment of equilibrium and expressed as
g/g dry lms. All the experiments were replicated three times and
the average values were inserted in the model developed by Singh
and Kulshrestha [25]:
1 1 1
+ (3)
(m m0 ) k(me m0 ) (me m0 )
(1)
where m0 and me are masses of the dry and fully equilibrated lm
samples respectively; and m is the mass of the hydrated lm at
different time intervals.
 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247
Fig. 1. In-lab built apparatus for VIPI approach.
2.8. Water vapor permeation studies
A modied ASTM standard (inverted-cup, E96-90, procedure
D) was employed to determine the water vapor transmission rate
(WVTR) as described elsewhere [1]. In brief, disposable cup, id
30 mm, was lled with 5 g of de-ionized water and test lm sample
(pre-conditioned at a relative humidity environment of 40%) was
placed on its mouth and xed using a cello tape (Kores, India). The
effective area of lm was approximately 7.5 104 m2 . The cup
was now placed in a desiccator lled with saturated solution of LiCl2
to provide a relative humidity (RH) of 20%, pre-maintained at 300 K.
The cup was taken out at regular time intervals, weighed accurately
using a digital electronic balance and MVTR was calculated using
the following expression:
24m
MVTR =
A t
where m = water loss (g); t = time period (h); A = effective transfer
area per m2 .
2.9. Expansion study
The expansion of wound dressing lm on wound surface was
mimicked by studying change in diameter of the circular lm sam-
ples in a 4% gelatin solution as described elsewhere [26]. In brief,
4 g of gelatin powder was dissolved in 100 mL of distilled water
at 85 C under moderate stirring till a clear solution was obtained.
Then, 30 mL of this solution was put into a Petri plate and allowed
to cool over night at 25 C. The lm sample with known diameter
was placed on the gelatin surface and change in its diameter was
monitored periodically till the sample attained a constant diameter.
The expansion ratio (ER) was expressed as
Diameter at time t(Dt )
ER =
Initial diameter (D0 )
2.10. Thrombogenicity and hemolytic potential study
The blood compatibility of sample Ch/CMC(20) was evaluated
according to procedure reported in the International Standard
Organization (ISO) [27] and two types of blood interactions were
studied i.e. thrombogenicity and hemolytic potential. All studies
were carried out in triplicate. The saline solution used for this study
was prepared by taking 0.9% NaCl, pH was adjusted to pH = 7.4
[28]. The % thrombosis and % hemolysis tests were performed as
described elsewhere [29]. A piece of lm with surface area of 4 cm2
was kept in25 mL of buffer saline (0.9%) for 3 h at 37 C. The saline
was drained out and 0.05 mL of acid citrate dextrose (ACD) blood
241
and after 10 min 0.03 mL of 0.1 M CaCl2 was placed on the surface
of the lm. The clotting was stopped after 10 min by adding 4 mL
of distilled water, followed by addition of 2 mL of formaldehyde to
x the clots. The clot was ltered, dried and nally weighted. The
positive and negative controls were taken as glass beaker without
sample and glass beaker without sample and blood. The thrombus
percentage was calculated as follow:
Mass of test sample Mass of() control
Thrombus (%) = 100
Mass of(+) control Mass of() control
The hemolysis tests were performed as described in American
Society for Testing and Materials (ASTM). Test sample Ch/CMC(20)
(area 1 cm2 ) was kept in 7 mL of 0.9% saline solution taken in test
tubes. After 24 h of incubation at 37 C, after that 0.9% saline was
drained out and 0.05 mL of acid citrate dextrose (ACD) blood were
(4) placed on the surface of each lm added to each sample take for
15 min. After 15 min add 10 mL 0.9% saline solution and kept incu-
bator at 37 C for 3 h. Positive and negative controls were prepared
by adding the same amount of ACD blood to 10 mL of distilled water
and 0.9% saline, respectively. Each tube was gently inverted twice
to make contact of the blood with the lm. After incubation, each
uid was transferred to a suitable tube and centrifuged at 104 rpm
for 15 min. The hemoglobin released by hemolysis was measured
by the optical densities (OD) of the supernatant at 545 nm using a
UVvisible spectrophotometer (Shimadzu, Genesis 10-S). The per-
centage of hemolysis was calculated as follows [30].
(OD of sample OD of(ve) control)
%Hemolysis = 100
OD of(+ve) control OD of(ve) control
2.11. Procedure of in vitro cytotoxicity test
An in vitro cytotoxic study test using Test on Extract method
(5)
was performed with test sample Ch/CMC(20) based on ISO 10993-
5, 2009. The sample was soaked in 5 mL physiological saline prior
to extraction. After soaking completely with saline, additional 1 mL
saline was added before dilution of extract. The extract was mixed
with various volumes of MEM2X medium to get 12.5, 25.0 and 50%
extract. Physiological saline without test material, processed simi-
lar to the test sample Ch/CMC(20), was considered as control. The
positive control was prepared by diluting phenol stock solution
(13 mg/mL) to 1.3 mg/mL with culture medium containing serum
(dilute phenol) where as negative control was prepared by incu-
bating 1.25 cm2 Ultra High Molecular Weight Poly(ethylene) with
1 mL of physiological saline at 50 C for 72 h. Different dilutions of
test sample extract, positive control and 100% extracts of negative
control, reagent control were placed on sub conuent monolayer
242 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247

Table 1 LambertBeers plot obtained for drug solutions of known concen-

Cells were examined microscopically and cellular responses were scored.
trations.
Grade Reactivity Condition of all cultures

0 None Discrete intra-cytoplasmatic granules, no cell

lysis, no reduction of cell growth
1 Slight Not more than 20% of the cells are round,
loosely attached and without
intra-cytoplasmatic granules, or show changes
in morphology; occasional lysed cells are
present; only slight growth inhibition
observable
2 Mild Not more than 20% of the cells are round,
devoid of intra-cytoplasmatic granules, no
extensive cell lysis; not more than 50% growth
inhibition observable
3 Moderate Not more than 70% of the cell layers contain
rounded cells or are lysed; cell layers not
completely destroyed, but more than 50%
growth inhibition observable
4 Severe Nearly complete or complete destruction of
the cell layers
of L-929 cells. After incubation of cells with extract of test sample
Ch/CMC(20) and control at 37 1 C for 2426 h, cell culture was
examined microscopically for cellular response. Cells were exam-
ined microscopically and cellular responses were scored as 0, 1, 2,
3 and 4 based on following Table 1.
3. Results and discussion
3.1. Need to develop new strategy
In preliminary investigations, we prepared 25 mL of a 2% (w/v)
solution of chitosan and added to 0.05 g of CMC (i.e. 10% of chi-
tosan weight) under mild stirring for 15 min to ensure complete
mixing. We observed that the suspension took almost 10 min to
settle down completely. On evaporation, we did not get uniformly
dispersed CMC loaded Ch lm. There was uneven distribution of
CMC in the lm obtained. This led us to think of using phase inver-
sion technique to prepare the lm. However, instead of using phase
inverting solution (PIS) (which is generally a non-solvent for the
lm material), we used NH3 vapors which could rapidly penetrate
in to the cellulose-dispersed chitosan solution and cause the pre-
cipitation to occur in a very short period of time. We were able to
cause precipitation in as short time as 30 s, and obtained chitosan
lm with a uniform distribution of CMC throughout.
3.2. Preparation of Ch and CMC/Ch lms

2.12. Preliminary curcumin release study
The pre-weighed piece of drug-loaded lm was placed in 25 mL
of release medium (i.e. physiological uid) at 37 C. After regu-
lar time-intervals, lm was transferred into fresh release medium,
and the amount of drug released was determined spectrophoto-
metrically at 526 nm. The quantity of drug was calculated using
In solvent phase inversion technique, the lm forming solution
is placed in a Petri plate and the non-solvent is poured in excess,
which results in immediate precipitation of the lm. To accelerate
this process, we used NH3 vapors for phase inversion. These vapors
of NH3 molecules penetrate into the dope and produce NH4 OH
with a sudden change in pH from acid to alkalinity. This causes
an immediate precipitation of the dissolved chitosan as a lm.

Fig. 2. FTIR spectrum of sample Ch/CMC(20).

 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247
Fig. 3. SEM images of (A) Ch/CMC(0) and (B) Ch/CMC(20) lms.
Fig. 4. AFM images of samples (A) Ch/CMC(0) and (B) Ch/CMC(20).
3.3. Characterization of lms
The FTIR spectrum of sample Ch/CMC(20), as shown in Fig. 2,
has a number of peaks almost common.
For example, hydrogen bonded OH stretching, overlapped
to the stretching vibrations of NH from chitosan [31]. A sharp
peak at 1658 cm1 corresponds to OH bending of adsorbed
water molecules, overlapped with stretching vibration of sec-
ondary amide ( C O) of chitosan. Similarly, C H stretching of
CH2 groupsat 2901 cm1 also appears due to both cellulose and
chitosan. In addition, H C H and O C-Hin-plane bending vibra-
tions at 1415 cm1 , C O stretching at 1049 cm1 , and C O C
stretch of ether linkage (1,4-B-d-glucoside) at 1155 cm1 also indi-
cate presence of both of the polysaccharides. Further, bending
vibrations of methylene and methyl groups are also visible at
1374 cm1 and 1416 cm1 [32]. Finally, the C O C, C C O and
C C H deformation modes and stretching vibrations in which the
motions of the C-5 and C-6 atoms are there, at 898 cm1 and C-
OH out-of plane bending mode around 675 cm1 also conrm the
presence of cellulose [33]. The stretching vibrations of protonated
amine of chitosan (C O) appears at 1574 cm1 . The surface mor-
phology of the lms was investigated by SEM and AFM analysis.
Fig. 3(a) and (b) show the SEM images obtained at 100 magni-
cations with a bar scale of 500 m. It can clearly be seen that
surface of the plain lm Ch/CMC(0) is fairly smooth where as a
dense but uniform dispersed array of cellulose micro crystals are
visible throughout the lm matrix in the lm sample Ch/CMC(20)
in Fig. 3(b). The cellulose micro crystals, though of varying sizes,
appear to have uniform density along the whole surface. The results
of AFM images of plain Ch/CMC(0) and cellulose loaded Ch/CMC(20)
243
lms are shown in Fig. 4(a) and (b) respectively. It can be seen that
Fig. 4(a) shows a at surface with layered texture. On the other hand
Fig. 4(b) shows surface texture of cellulose loaded lm. The surface
looks fairly rough with micrometer sized grooves throughout the
surface. These grooves vary in height of a few micrometers and are
attributable to the presence of cellulose crystals.
3.4. Swelling behavior of lms
The investigation of swelling capacity of wound dressing lms is
important. The lm absorbs a large quantity of uid from the exu-
dates, tries to keep the wound bed dry and simultaneously releases
the entrapped ingredient to prevent bacterial infection [34]. The
dynamic water uptake of lm samples Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20), and Ch/CMC(30) is shown in Fig. 5.
The results reveal that sample Ch/CMC(0) shows higher water
uptake as compared to the samples. The equilibrium SR values were
11.51, 10.08, 9.81 and 9.77 g/g respectively, thus indicating that as
the cellulose content increases, the water uptake decreases. The
reason is that cellulose crystals, present within the lm matrix, do
not interact with incoming water molecules because of presence of
intra molecular H-bonding interactions [35]. In our previous work,
we observed the similar behavior for cellulose nanocrystals loaded
poly(sodium acrylate) lms [36]. The swelling behavior of hydrogel
matrix is described by several mathematical models [37]. According
to power function law which describes fractional water uptake as
a time-dependent phenomenon [38]:
Mt
= k tn (6)
M
244 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247

Fig. 5. Dynamic water uptake data for various samples Ch/CMC(0), Ch/CMC(10), Fig. 7. Schott model plots for lm samples Ch/CMC(0), Ch/CMC(10), Ch/CMC(20)
Ch/CMC(20) and Ch/CMC(30) in PF at 37 C. and Ch/CMC(30) in PF at 37 C.

where Mt and M are the masses of the swollen hydrogel at time t
and in the initial dry state respectively; n and k are swelling expo-
nent and gel characteristic constant respectively. The logarithmic
form of Eq. (5) is:
Mt
ln F = ln k + n ln t
M of above equation yields:
The dynamic water uptake data were used to obtain linear plots
between ln F and ln t (data not shown) and their slopes and inter-
cepts enabled to evaluate n and k. The values of n and k for
the samples Ch/CMC(0), Ch/CMC(10), Ch/CMC(20) and Ch/CMC(30)
were found to be 0.483 and 16.70 102 ; 0.462 and 10.31 102 ,
0.310 and 19.9 102 , 0.369 and 0.168 102 respectively. It is
noted that for all the samples, n lies below 0.5, indicating that rate
of water penetration is much below the rate of relaxation of poly-
meric chains. This is specically termed as less-Fickian diffusion
[39]. The reason is that at pH 7.4, the chitosan chains do not carry
charge on their amino groups, and therefore chain relaxation type
process is non-operative. In addition, the cellulose crystals act as
diffusion barrier for incoming water molecules. The mutual entan-
glements between un-crosslinked chitosan chains and cellulose
micro crystals also contribute to this non-relaxational less-Fickian
behavior. We also applied the First order kinetic model, according
to which the rate of water uptake is given as:
dMt
= k1 (M Mt ) (8)
dt
where k1 is the rst order rate constant for swelling. On integrating
above equation with the limits t = 0, Mt = 0; and t = t, Mt = Mt , we get,
1 Mt
ln = k001 t (9)
M imental values of M differ appreciably. It is because the two
Thus, ln(1 Mt/ M ) versus t plots should be linear with a slope
equal to k1 (see Fig. 6). The slope of the plots yielded constants
k1 , as given in Table 2. According to Schott model; swelling rate at
Fig. 6. First order kinetic plots for various samples Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20) and Ch/CMC(30) in PF at 37 C.
any time is directly proportional to the quadratic of the swelling
capacity before the attainment of equilibrium state [40]
dMt
= k2 (M Mt )2 (10)
dt
where k2  is the second order rate constant for swelling. Integration
(7)
t 1 t
= 2
+
Mt k2 M M
or
t
= A + Bt (11)
Mt
where A and B are coefcients with following signicance: at a long
retention time Bt  A and therefore B = 1/M . On the contrast, at a
very short time interval Bt  A and so,
dMt 1
Lt =
dt A
t0
Therefore, the intercept A is reciprocal of initial swelling rate.
Finally the Schott kinetic rate constant k2 is calculated as:
slope2 B2
k2 = = (12)
Intercept A
The t/Mt versus t plots for the samples Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20) and Ch/CMC(30) are shown in Fig. 7. The various
related parameters are given in Table 2.
It is clear that for all the samples, the theoretical and exper-
components of these lms, i.e. cellulose and chitosan have dif-
ferent afnity for solvent molecules. However, a fair agreement
between these values in the case of homo-polymer or co-polymer
with almost comparable afnity toward solvent molecules is
reported. In a study [41] ESR(theo) and ESR(exp) for carboxymethyl
cellulose-g-poly(acrylic acid-co-2-acrylamido-2-methyl propane
sulfonic acid) hydrogels were almost same, i.e. 465.6 versus 462.9,
455.8 versus 452.4, 386.4 versus 390.6. Since CMC, acrylic acid,
and 2-acrylamido-2-methyl propane sulfonic acid, are pH sen-
sitive and hydrophilic in nature, they show similar afnity for
water. It is also noticeable in Table 2 that k1 and k2 lie in the
usual range of 18.5 103 to 25.6 103 min1 and 1.5 103
to 5.2 103 (g/g min)1 respectively. In a study, Karadag et al.
[40] have also reported k1 and k2 values for poly(acrylamide-co-
crotonic acid) hydrogels in the range of 3.5 to 39.4 103 min1
and 0.7 to 2.1 103 (g/g min)1 respectively. Similarly, k2 val-
ues for dimethyl amino ethyl acrylate methyl chloride quaternary
salt based hydrogels were in the range of 4.8 to 37.6 103 min1
 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247 245

Table 2
Parameters associated with various kinetic models.

Sample code Diffusion model First order model Schott model

kd (min)1/2 R2 k1 (min1 ) R2 k2 103 (g/g min)1 ESR(theo) (g/g) ESR(exp) (g/g) R2

Ch/CMC(0) 0.0874 0.9948 0.0205 0.9918 1.5 14.28 11.57 0.9958

Ch/CMC(10) 0.0787 0.9855 0.0185 0.9678 1.6 12.50 10.14 0.9849
Ch/CMC(20) 0.0814 0.9243 0.0221 0.8995 5.2 11.11 9.81 0.9834
Ch/CMC(30) 0.0932 0.9594 0.0256 0.9544 3.6 9.09 7.78 0.9892

Fig. 8. Dynamic moisture uptake for the samples Ch/CMC(0), Ch/CMC(10),

Ch/CMC(20) and Ch/CMC(30). Fig. 10. Kinetics of moisture vapor permeation for the samples Ch/CMC(0),
Ch/CMC(10), Ch/CMC(20) and Ch/CMC(30).

[42]. Finally, the order of tness was; Schottmodel > Diffusion

model > First order model.
3.5. Dynamic moisture adsorption studies
The dynamic moisture uptake by lm samples was investigated
under the RH environment of 95% at 30 C. The lm samples exhibit
water vapor sorption curves (see Fig. 8) as obtained for polysaccha-
rides based lms [40].
The samples Ch/CMC(0), Ch/CMC(10), Ch/CMC(20), and
Ch/CMC(30) show a decrease in moisture uptake with cellulose
contents. This is because the number of capillaries existing on the
lm surface decreases with CMC content, and therefore moisture
penetration into the lm also decreases. The equilibrium moisture
content (EMC) for these samples was 0.105, 0.092, 0.075 and
0.062 g/g respectively. One more reason for observed decrease is
that with increase in the CMC content, the lm matrix becomes
more compact, thus putting hindrance toward incoming moisture.
In order to determine the theoretical (EMC)theor , KulshresthaSingh
model was used. Plots between 1/(mt m0 ) and 1/t which yielded
straight lines (see Fig. 9) whose slopes and intercepts were used to
determine (EMC)theor .
Fig. 9. 1/(mt m0 ) versus 1/t plots for the samples Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20) and Ch/CMC(30).
It was found that for the samples Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20), and Ch/CMC(30), the (EMC)theor and (EMC)exp val-
ues were 0.091 versus 0.105, 0.083 versus 0.092, 0.065 versus
0.075 and 0.052 versus 0.062 respectively, indicating not a fair
agreement. Unlike this work, in a study by Rhim and Wang [43]
there was close agreement between the theoretical and exper-
imental values for agar/carrageenan/konjacglucomannan ternary
blends.
3.6. Moisture permeations study
A good dressing keeps the wound environment moist and pos-
sesses an appropriate moisture vapor transmission rate (MVTR). A
higher MVTR may lead to dehydration of wound and adherence of
the dressing to wound bed [44]. Contrary to this, a low MVTR may
cause leakage of exudates from the edges of lm, thus promoting
bacterial infection [45]. The results of permeation test are shown
in Fig. 10.
It is clear that MVTR decreases with increase in the CMC
content. The MVTR values, determined using Eq. (2), and were
8543, 4580, 7206 and 5241 g1 /m2 /day for the samples Ch/CMC(0),
Ch/CMC(10), Ch/CMC(20) and Ch/CMC(30) respectively. The MVTR
of the normal skin is about 204 g/m2 /day. However, for injured skin
it can range from 260 to 5400 g/m2 /day [46]. A wound dressing
with suitable MVTR prevents excessive dehydration and buildup
of exudates. According to reports available regarding the suitabil-
ity of wound dressing lm on the basis of their MVTR, it has been
recommended that normal MVTR for dressings range from 70 to
9400 g/m2 /day [47]. This range has been set up on the basis of
condition of wounds such as low injured skin, rst degree burns,
granulating wounds etc. It can be seen that MVTR of all the sam-
ples lie in the range of 709400 g1 /m2 /day which is the desirable
range for most of the wound dressings. Dias et al. [1] reported
a MVTR of carboxybutyl chitosan and agarose based lms in the
range of 35974352 g/m2 /day. However, there may be variations
depending upon the conditions like: humidity, temperature, etc.
It is concluded that MVTR of all lms fall within the prescribed
range.
246 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247
Fig. 11. Expansion ratio (ER) versus time plots for the samples Ch/CMC(0),
Ch/CMC(10), Ch/CMC(20) and Ch/CMC(30).
3.7. Expansion study
The gelatin solution is used to mimic a supporting wound. The
study of expansion behavior of a lm on the surface of gelatin
medium provides useful information regarding suitability of the
lm in high exudate wounds. The results of expansion study are
shown in Fig. 11 for the lm samples Ch/CMC(0), Ch/CMC(10),
Ch/CMC(20) and Ch/CMC(30).
It can be seen that the sample Ch/CMC(0) undergoes a faster
expansion in diameter and after 3 h it acquires an expansion ratio
of 2, whereas the sample Ch/CMC(10) shows a relatively slower
increase in the expansion ratio(ER) and after 3 h it acquires an ER
of 1.3. It was observed that both of the lm samples retained their
circular shape even after 24 h. The results are in agreement with
the water absorption behavior of these two lms as discussed pre-
viously. Similarly, the sample Ch/CM (20) undergoes almost similar
changes in its diameter and exhibit an ER of 2.0. However, the sam-
ple Ch/CMC(30), containing maximum amount of cellulose micro
crystals shows a much slower increase in the diameter and after
60 min it attains maximum ER of 1.5. The reason is that high CMC
content does not permit the lm to undergo larger expansion. It is
Fig. 12. (a) In vitro cytotoxic study for the sample Ch/CMC(20) of 12.5% extract (b) 25% extract (c) 50% extract.
noteworthy that all the samples retained their shape stability even
after 36 h. Similar results are reported by [48] for alginate based
single layer lm. Same authors reported that early transformation
of a lm texture into gel may leave the lm with gummy residue
deposited around the wound and hence it may be painful to the
patient to replace the dressing. In the light of this fact it appears
that the lm samples Ch/CMC(0), Ch/CMC(10), and Ch/CMC(20),
may be more suitable.
3.8. Thrombus formation and hemolysis
The % thrombus for the lm sample Ch/CMC(20) was 11.1%
which falls well within the prescribed range of 1020%, thus indi-
cating that the lm sample was fairly non-thrombogenic. Finally,
the % hemolysis was calculated to be 1.8%, thus suggesting a non-
hemolytic lm material.
3.9. Cytotoxicity of Ch/CMC(20) lm
As per ISO 10993-5, the achievement of numerical grade more
than 2 is considered as cytotoxic. The results for 12.5%, 25% and 50%
extracts are shown in Fig. 12(a)(c) respectively.
It can be seen that there are discrete intra cytoplasmatic gran-
ules, no cell lysis and there is no reduction of cell growth. This
indicates that the lm sample Ch/CMC(20) is absolutely non-
cytotoxic.
3.10. Release of CC from the lms
The release proles of lm samples Ch/CMC(0)0.18 and
Ch/CMC(20)0.18 in physiological uid (PF) at 37 C are shown in
Fig. 13.
It is seen that the sample Ch/CMC(0)0.18 demonstrates a faster
release as compared to the sample, Ch/CMC(20)0.18 which may sim-
ply be due to the fact that cellulose micro crystals, present within
the lm matrix, act as diffusion barrier and put hindrance toward
the release of CC molecules. The total release demonstrated by the
two samples was 0.11 and 0.06 g/g lm respectively. Therefore, it
 S.K. Bajpai et al. / International Journal of Biological Macromolecules 75 (2015) 239247
Fig. 13. CC release proles for the samples Ch/CMC(0)0.18 and Ch/CMC(20)0.18 in PF
at 37 C.
can be inferred that variation in cellulose concentration within the
lm matrix could be an effective parameter to obtain the curcumin
release at a pre-determined rate.
4. Conclusion
This study concludes that VIPI technique can be successfully
employed to obtain a uniform distribution of cellulose micro crys-
tals within the chitosan lm. The water absorption and moisture
permeation properties of the lm can be controlled by variation
in CMC concentration in the lm matrix. The lm may success-
fully be employed to obtain controlled release of a bioactive agent
from the lm for wound healing. Here, it is worth mentioning
that in order to obtain controlled release from the micro crys-
talline cellulose/chitosan matrix it is also required to focus on the
microstructure of cross-section of the various lms synthesized.
Moreover, a study of distribution of cellulose crystals within the
matrix and their orientation is also needed to obtain correlation
between the controlled release and microstructure of dispersed
cellulose.
Conict of interest
All the authors hereby declare that we do not have any conict
of interest about this manuscript.
Acknowledgment
The authors are thankful to IIITDM, Jabalpur, India for providing
facilities for SEM and AFM analysis.
References
[1] A.M.A. Dias, M.E.M. Braga, I.J. Seabra, P.M. Ferreira, H. Giall, H.C. de Sousa, Int.
J. Pharm. 408 (2011) 919.
[2] B. Singh, L. Pal, Mech. Behav. Biomed. Mater. 9 (2012) 921.
[3] H. Xu, L. Ma, H. Shi, C. Gao, C. Han, Polymer 18 (2007) 869875.
247
[4] S.B. Patil, V.S. Mastiholimath, A.R. Kulkarni, Orient. Pharm. Exp. Med. 11 (2011)
123129.
[5] D. Archana, B.K. Singh, J. Dutta, P.K. Dutta, Carbohydr. Polym. 95 (1) (2013)
530539.
[6] C.J. Doillon, C. Deblois, M.F. Fournier, N. Cote, Mater. Sci. Eng. 2 (1994) 4347.
[7] J.K. In-Yong, S. Seo, S.H. Moon, M.K. Yoo, B.C. Kim, C.S. Cho, Biotechnol. Adv. 26
(2008) 121.
[8] G. Saraswathy, S. Pal, C. Rose, P.T. Sastry, Bull. Mater. Sci. 24 (2001) 415420.
[9] A. Anitha, N. Deepa, P.K. Chennazhi, M.K. Yoo, P. In-Young, B.C. Kim, C.S. Cho,
Carbohydr. Polym. 83 (2011) 6673.
[10] S. Senel, S.J.M. Clure, Adv. Drug Deliv. Rev. 5 (2004) 14671480.
[11] R. Jayakumar, R.L. Reis, J.F. Mano, Drug Deliv. 14 (2007) 917.
[12] M.S.M. Eldin, E.A. Soliman, A.I. Hashem, T.M. Tamer, Trends Biomater. Artif.
Organs 22 (2008) 154168.
[13] L. Ma-Hock, P.M. Farias, T. Hofmann, A.C. Andrade, J.N. Silva, T.M. Arnaud, W.
Wohlleben, V. Strauss, S. Treumann, C.R. Chaves, S. Grters, R. Landsiedel, B.
van Ravenzwaay, Toxicol. Lett. 225 (2014) 2026.
[14] P.R. Umashankar, P.V. Mohanan, T.V. Kumari, Toxicol. Int. 19 (1) (2012) 558.
[15] J. Jin, M. Song, D.J. Hourston, Biomacromolecules 5 (1) (2004) 162168.
[16] C. Chen, Z. Gao, X. Qiu, S. Hu, Molecules 18 (6) (2013) 72397252.
[17] A.H. Jawad, M.A. Nawi, Polymers 90 (1) (2012) 8794.
[18] A.P. Vieira, S. Badshah, C. Airoldi, Int. J. Biol. Macromol. 52 (2013) 107111.
[19] V.C. Nguyen, V.B. Nguen, M.-F. Hsieh, Int. J. Polym. Sci. 106570 (2013) 17.
[20] F. Mang, X.H. Fan, J. Biomed. Mater. Res. Part B: Appl. Biomater. 102 (3) (2014)
533542.
[21] A.K. Popuri, B. Pagala, Int. J. Innov. Res. Stud. 2 (5) (2013) 199289.
[22] S. Revathy, S. Elumalai, M. Benny, B. Antony, J. Exp. Sci. 2 (7) (2011) 2125.
[23] N. Saraswathi, R. Rohit, K. Shanmugam, S. Charanya Sozseehwari, P. Rama-
lingam, Indian J. Nat. Prod. Resour. 3 (2) (2012) 237241.
[24] E. Ayranci, O. Duman, J. Food Eng. 70 (2005) 8391.
[25] B.P.N. Singh, S.P. Kulshrestha, J. Food Sci. 52 (1987) 15381541.
[26] K.H. Matthews, H.N.E. Stevens, A.D. Auffret, M.J. Humphrey, G.M. Eccleston, Int.
J. Pharm. 289 (2005) 5162.
[27] Geneva International Standard Organization, ISO, Biological Evaluation Of Med-
ical DevicesPart 4: Selection of Tests for Interaction with Blood, 10993-4,
1999.
[28] Y. Lai, W. Yin, J. Liu, R. Xi, J. Zhan, Nanoscale Res. Lett. 5 (2010) 302307.
[29] M.D. Rockville, US Pharmacopeia XXIII, 119, US Pharmacopeial Convention,
1994.
[30] A. Chhatri, A.K. Bajpai, S.S. Shandhu, N. Jain, J. Biswas, Polymers 83 (2011)
876882.
[31] C. Paluszkiewicz, E. Stodolak, M. Hasik, M. Blazewicz, Spectrochim. Acta Part A:
Mol. Biomol. 79 (2011) 784788.
[32] M. Darder, M. Colilla, E. Ruiz-hitzky, Chem. Mater. 15 (2003) 37743780.
[33] J.X. Sun, X.F. Sun, H. Zhao, R.C. Sun, Polym. Degrad. Stab. 84 (2004) 331339.
[34] K. Kurhade, M. Momin, P. Khanekar, S. Mhatre, Int. J. Drug Deliv. 5 (2013)
353361.
[35] A. Isogai, R.H. Atalla, Cellulose 5 (1998) 309319.
[36] S.K. Bajpai, V. Pathak, B. Soni, Y.M. Mohan, Carbohydr. Polym. 106 (2014)
351358.
[37] A. Blanco, G. Gonjalez, E. Casanova, M.E. Pirela, A. Briceno, Appl. Math. 4 (2013)
161170.
[38] N.A. Peppas, R.W. Korsmeyer, Hydrogels in Medicine and Pharmacy, CRC Press,
Boca Raton, 1987.
[39] F. Ganji, S. Vasheghani-Farahani, E. Vasheghani-Farahani, Iran. Polym. J. 19
(2010) 375398.
[40] E. Karadag, D. Saraydin, N. Sahiner, O. Guven, J. Macromol. Sci. Part A: Pure Appl.
Chem. 38 (11) (2001) 11051121.
[41] Y. Wang, X. Shi, W. Wang, A. Wang, Turk. J. Chem. 37 (2013) 149159.
[42] I. Katime, E. Mendizabal, Mater. Sci. Appl. 1 (2010) 162.
[43] J.-W. Rhim, L.-F. Wang, Carbohydr. Polym. 96 (2013) 7181.
[44] J.N. Kim, J.W. Rhim, Food Sci. Biotechnol. 13 (2004) 616621.
[45] Y. Hu, V. Topolkaraev, A. Hiltner, E. Baer, J. Appl. Polym. Sci. 81 (2001)
16241633.
[46] L.G. Ovington, Dermatology 25 (2007) 3338.
[47] F.L. Mi, Y. Wu, S.S. Shyu, J.Y. Schoung, Y.B. Huang, Y.H. Tsai, J.Y. Hao, J. Biomed.
Mater. Res. 59 (2002) 438449.
[48] H. Thu, H.M. Zulfakar, Int. J. Pharm. 434 (2012) 375383.