Journal of Controlled Release 96 (2004) 9 20

www.elsevier.com/locate/jconrel

Poly(vinyl alcohol) and poly(acrylic acid) sequential

interpenetrating network pH-sensitive microspheres
$
for the delivery of diclofenac sodium to the intestine
Mahaveer D. Kurkuri, Tejraj M. Aminabhavi *
Drug Delivery Division, Center of Excellence in Polymer Science, Karnatak University, Dharwad 580 003, India
Received 11 May 2003; accepted 18 December 2003

Abstract
Sequential interpenetrating network (IPN) of poly(vinyl alcohol) (PVA) and poly(acrylic acid) (PAA) were prepared and
crosslinked with glutaraldehyde (GA) to form pH-sensitive microspheres by the water-in-oil (w/o) emulsification method.
Microspheres were used to deliver a model anti-inflammatory drug, diclofenac sodium (DS), to the intestine. The formed IPN
was analyzed by Fourier transform infrared spectroscopy (FTIR). Differential scanning calorimetry (DSC) and X-ray diffraction
(XRD) analyses were done on the drug-loaded microspheres to confirm the polymorphism of DS. Results indicated a molecular
level dispersion of DS in the IPN. Microspheres formed were spherical with the smooth surfaces as evidenced by scanning
electron microscopy (SEM). Particle size and size distribution was studied using laser light diffraction particle size analyzer.
Particle size analysis was also done by optical microscope for the selected microspheres; the change in diameter of the
microspheres when soaked in different media at different time intervals was measured by optical microscope. Microspheres
showed a pulsatile swelling behavior when the pH of the swelling media was changed. The swelling data were fitted to an
empirical equation to understand the phenomenon of water transport as well as to calculate the diffusion coefficient (D). Values
of D in acidic media were lower than those found in basic media. The values of D decrease with increasing crosslinking of the
matrix. In-vitro release studies have been performed in 1.2 and 7.4 pH media to simulate gastric and intestinal conditions. The
results indicated a dependence on the pH of the release media, extent of crosslinking and the amount of drug loading.
D 2004 Elsevier B.V. All rights reserved.
Keywords: Microspheres; pH-sensitive; Sequential IPNs; Poly(vinyl alcohol); Anti-inflammatory

1. Introduction
Oral controlled release multiple unit dosage forms
such as beads, pellets and microspheres are becoming
more popular than single unit dosage forms due to

This article is CEPS Communication No. 26.

* Corresponding author. Fax: +91-836-2771275.
E-mail address: aminabhavi@yahoo.com (T.M. Aminabhavi).
0168-3659/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jconrel.2003.12.025

several inherent advantages [1 3]. The utility of such

products can be further extended if the system is
responding to the surrounding environment. The pHsensitive dosage forms for oral delivery are therefore
attractive alternative systems for the enteric coating to
deliver acid-labile drugs [4] and polypeptides [5]. It is
well known that hydrogels containing typical chemical groups can respond to external stimuli such as pH,
ionic strength, temperature and electric current [6,7].
The pH-sensitivity of hydrogels is due to the presence

10

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

of weakly acidic and/or basic functional groups on the

polymer backbone. Their water-uptake properties are
attributed to the ionization of functional groups, which
depends upon the pH and ionic strength of the
external medium where the hydrogel is placed, thus
making the system pH-sensitive. Particularly, synthetic polymers like poly(methyl methacrylate) [8], poly
(acrylic acid) [9], poly(N,N-iso-propylacrylamide) [9],
and natural polysaccharide such as chitosan [10] have
been used as pH-sensitive drug delivery systems.
Many attempts have been made in the earlier
literature to develop pH-sensitive matrix systems for
the oral delivery of drugs. Poly(acrylic acid) (PAA)
has been used as a side chain on the main poly(vinyl
alcohol) (PVA) backbone to produce pH-sensitive
hydrogels by in situ hydrolysis of acrylamide [11] or
can be grafted onto other polymers such as poly
(vinylidene fluoride) [12] by employing tedious reaction conditions involving preirradiation [13]. In the
present research, an easy and direct method has been
adapted to incorporate PAA into PVA backbone by the
formation of a sequential interpenetrating network
(IPN) to obtain pH-sensitive polymeric matrix.
Multiple unit dosage forms distribute more uniformly in the gastrointestinal tract (GIT), resulting in
a more uniform drug absorption and reduced local
irritation when compared to single unit dosage forms
containing highly acidic drugs such as diclofenac
sodium. High local drug concentration and the risk
of toxicity due to locally restrained tablet can be
avoided with the multiple unit dosage forms. The
advantages of such controlled release (CR) preparations containing non-steroidal anti-inflammatory
drugs (NSAID) over their conventional dosage forms
have been reported earlier by Kulkarni et al. [14].
Particularly, such systems minimize the serious gastric irritant side effects of the conventional NSAID
preparations [15].
In an earlier study, we have prepared [16,17] the
stimuli-responsive polymeric matrices to deliver drugs
to the intestine, colon, etc. The present investigation is
an extension of these efforts to prepare microspheres
from the sequential IPNs of PVA and PAA to produce
pH-sensitive hydrogels due to the presence of COOH
groups in the microspheres. Microspheres with three
different PAA compositions and three different extent
of crosslinking were prepared. These microspheres
were employed for loading NSAID, diclofenac sodi-

um (DS) in three different amounts. The in vitro drug

release was investigated in gastric and intestinal pH
conditions (i.e., 1.2 and 7.4 pH, respectively). The
matrices developed are particularly useful as pHsensitive drug release devices to the intestine.

2. Materials and methods

2.1. Materials
PVA having mol. wt. of 125,000, acrylic acid, ceric
ammonium nitrate (CAN), Tween-80, glutaraldehyde
(GA) (25% w/v), hydrochloric acid, light liquid paraffin and hexane used in this study were procured
from s.d. Fine Chemicals, Mumbai, India. A gift
sample of DS was obtained from Bioethicals Pharmaceutical (Hubli, India). Doubly distilled water was
used throughout the study.
2.2. Preparation of IPN and drug loading
About 4 g of PVA was dissolved in 100 ml distilled
water. Different solutions containing 10, 20 and 30
mass% of PAA were prepared. Required amount of
acrylic acid monomer was taken in 10 ml water, which
was then added drop-wise to PVA solution in a round
bottom flask maintained at 60 jC. Then, 0.2 g of CAN
dissolved in 10 ml of water was added to the above
solution, stirred vigorously until the above solution
reduced to about 50 ml. Then the solution was cooled
to room temperature. Different mass% of DS (i.e., 5%,
10%, and 20%) was loaded by dissolving in 10 ml
methanol in a separate beaker and then, adding it
slowly to the above polymer solution by heating
gently in order to avoid reprecipitation.
The above prepared solution was emulsified into
liquid paraffin to form water-in-oil (w/o) emulsion at
400 rpm using Eurostar (IKA Labortechnik, Germany) for 30 min in a separate 1000 ml beaker
containing 100 ml of light liquid paraffin, 2% (w/v)
of Tween-80, 1 ml of 0.1 M HCl and the required
amount of GA. Microgels with different extents of
crosslinking were prepared by taking 2.5, 5.0 and 7.5
ml of GA. The microspheres formed were filtered and
washed repeatedly with hexane and water to remove
any excess amount of surfactant, unreacted CAN as
well as GA. These microspheres were then dried

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

11

under vacuum at 40 jC and stored in a desiccator

before further analytical testing.
In total, 12 formulations were prepared by varying
three parameters, i.e., amount of PAA, extent of drug
loading and extent of crosslinking. In order to understand the variables, formulation codes are assigned as
given in Table 1. For instance, the formulation code, Fabc, refers to three variables viz., arepresenting the
amount of PAA numbered as 0, 1, 2 and 3 for 0%, 10%,
20% and 30% PAA in the IPN, brepresents three
drug loadings numbered as 1%, 2% and 3% for 5%,
10% and 20% of drug in the microspheres and c
represents three extents of crosslinking numbered as 1,
2 and 3 for 2.5, 5 and 7.5 ml of GA added.

amounts of PAA. The temperature was increased at

the rate of 20 jC/min up to 350 jC. These analyses
were done on a Perkin Elmer, DSC-II (USA) available
at Indian Institute of Science, Bangalore (courtesy of
Mr. Eshwar Jarali).

2.3. Fourier transform infrared spectra (FTIR)

2.6. Scanning electron microscopic (SEM) analysis

Polymer samples or microspheres were crushed to

make KBr pellets under a hydraulic pressure of 600
kg/cm2. FTIR (Nicolet, Model Impact 410, USA)
spectra were taken in the wavelength region between
400 and 4000 cm
1.

A few samples of microspheres were taken on

copper stub and sputtered with gold for 2 min.
These gold-coated microspheres were mounted on
the SEM instrument (Leica 400, Cambridge, UK)
and photographs were taken at magnification 100 
and 300 .

2.5. X-ray diffraction (XRD) studies

XRD curves were obtained for pure DS and the
DS-loaded microspheres containing three different
amounts of PAA, i.e., 10%, 20% and 30%. The 2h
was increased up to 38j.These analyses were done
with x-pert, Philips (UK) at Indian Institute of Science, Bangalore (courtesy of Mr. Eshwar Jarali).

2.4. Differential scanning calorimetric (DSC) studies

2.7. Particle size analysis
DSC analyses were performed for pure DS and
drug-loaded microspheres of IPN containing different

Table 1
Results of percent encapsulation efficiency and mean size of the
microspheres at 5 ml GA and stirring speed of 400 rpm
Formulation % PAA in
%
codea
microspheres Diclofenac
sodium
loaded

%
Encapsulation
efficiency
F S.D.

Mean
particle
size
(Am) F S.D.

F-012
F-022
F-032
F-112
F-122
F-132
F-212
F-222
F-232
F-312
F-322
F-332

51.0 F 1.3
53.0 F 2.1
58.0 F 0.8
71.9 F 1.2
72.0 F 1.8
77.9 F 0.6
80.6 F 0.6
82.0 F 0.9
82.6 F 1.2
86.5 F 0.7
90.8 F 0.8
91.1 F 1.4

160 F 1.2
176 F 1.3
192 F 0.9
224 F 0.8
240 F 1.5
256 F 1.3
256 F 1.4
272 F 0.9
272 F 0.8
288 F 0.8
288 F 1.2
304 F 0.9

10
10
10
20
20
20
30
30
30

5
10
20
5
10
20
5
10
20
5
10
20

a
F-112 refers to formulation with three parameters, three PAA
compositions, three DS loadings and three crosslinking amounts.

Particle size of the microspheres was measured

by using a laser light scattering particle size analyzer
(Mastersizer 2000, Malvern Instruments, UK).
About 500 mg of microspheres were suspended in
100 ml distilled water containing 0.1% NaCl. This
suspension was transferred to wet sample holder and
the suspension was stirred under sonication to avoid
agglomeration of particles during measurements. For
measurement of sizes of different formulations/
batches, the sample holder was cleaned with distilled water followed by acetone to prevent cross
contamination.
2.8. Swelling study
The pH-dependent equilibrium swelling of the
empty microspheres and the DS-loaded crosslinked
microspheres were studied both in the simulated
gastric and intestinal pH conditions using 0.1 N
HCl and 7.4 pH phosphate buffer, respectively.
Microspheres were allowed to swell completely
for about 24 h to attain equilibrium at 37 jC.

12

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

Adhered liquid droplets on the surface of the

particles were removed by blotting with tissue
papers and the swollen microspheres were weighed
on an electronic balance (Mettler, Model AT 20,
Switzerland). The microspheres were dried in an
oven at 60 jC for 5 h until there was no change in
the dry mass of the samples. From the equilibrium
mass%, Wl of the sample, water uptake, Q was
calculated by measuring the dry mass, W0 using the
equation:

3. Results and discussion

3.1. FTIR spectroscopic study

Dynamic swelling of the microspheres was measured by using a light microscope. The changes in
diameter were monitored precisely with an ocular
microscope under the plane-polarized light at room
temperature in gastric and intestinal pH conditions.
Liquid droplets were removed by using blotting
papers and again fresh media was added; this step
was repeated for studying the pulsatile swelling of
the microspheres.

Crosslinked polymeric IPNs of PAA and PVA were

washed repeatedly with water to remove the unbound
PAA from the matrix. Hence, FTIR peaks due to
carboxylic acid groups of PAA in the polymer network confirm the presence of PAA. Microspheres
were scanned in the infrared range between 4000
and 400 cm
1 using KBr pellets. FTIR spectra of
the microspheres with different extent of crosslinking
are shown in Fig. 1. An increase in peak intensity
appeared at 1022 cm
1 due to an increase in extent of
crosslinking is assigned to the formation of acetal ring
and ether linkage as a result of the reaction between
hydroxyl groups of PVA and aldehydic groups of
glutaraldehyde. The peak at 1133 cm
1 is attributed
to C O stretching mode in PAA, while the peak at
819 cm
1 is due to the O H out of plane motion of
the carboxylic group in PAA [18,19]. Peak intensities
of carbonyl group and CMO stretch increase gradually, while the peak intensity of O H out of plane
motion decreases.

2.9. Drug loading

3.2. Differential scanning calorimetric study

Amount of DS loaded in the microspheres was

estimated by crushing 50 mg of the swollen microspheres in 10 ml of 7.4 pH phosphate buffer at 50 jC to
extract drug from the microspheres. The solution was
centrifuged to remove the suspended polymer particles
and the clear supernatant liquid was diluted with the
buffer solution. Drug was assayed using the UV VIS
spectrophotometer (Model Anthelie, Secomam,
France) at k max of 277 nm.

DSC graphs of pure DS (curve a) and DS-loaded

IPN microspheres containing 10, 20 and 30% PAA
(curve b, c and d, respectively) are presented in Fig. 2.
The DS shows sharp peaks at 108 jC and 252 jC;
these peaks are due to polymorphism and melting of
DS. DSC graphs of DS-loaded microspheres of different IPN show blunt peaks in the range of 127 to
157 jC due to the Tg of IPNs. As the content of PAA
in IPN increases, the Tg of IPN has shifted to higher
temperature indicating an increased rigidity of IPN.
All the drug-loaded microspheres did not show any
peak at 252 jC as observed in curve a (DS). This
confirms that DS is molecularly dispersed into IPNs
and hence, no separate peak is observed.

Q

Wl
W0
W0


 100

2.10. In-vitro drug release

In-vitro drug release was carried out at 37 jC in
a USP-II rotating paddle dissolution test apparatus
(Disotest LabIndia, Mumbai) at the rotation speed of
100 rpm. Drug release from the microspheres was
studied both in the simulated gastric (0.1 N HCl)
and intestinal (7.4 pH phosphate buffer) fluids. At
regular intervals of time, aliquot samples were
withdrawn and analyzed for drugs using the UV
VIS spectrophotometer.

3.3. X-ray diffraction (XRD) studies

XRD data of pure DS, DS-loaded IPN microspheres containing 10, 20 and 30% PAA were
obtained. The DS has shown highly intense sharp
peaks at around 2h of 20j, but the drug-loaded

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

13

Fig. 1. FTIR spectra of formulations containing 10% PAA without drug and crosslinked with 2.5 (curve A), 5 (curve B) and 7.5 ml GA (curve C).

microspheres have shown peaks at around 2h of 10j

to 15j. However, peaks due to DS crystals are not
shown in the drug-loaded microspheres confirming
that the drug is molecularly dispersed into the polymer
and no crystals of drug were found.
3.4. Scanning electron microscopic studies
SEM photograph of a single microsphere taken at
100  and 300  magnification is shown in Fig. 3.
Microspheres are spherical without forming agglomerations (Fig. 3A). The surface of microspheres is smooth
without any pores (Fig. 3B). Polymeric debris is seen
around some particles, which are due to the typical
method of particle production (i.e., simultaneous particle production and formation of IPN). Microspheres

produced using different ratios of PAA content did not

show any effect on the surface properties.
3.5. Particle size by laser light scattering
Results of mean particle size with standard errors
are presented in Table 1, while the size distribution
curve for typical formulation containing 10% PAA,
10% DS and 5 ml of GA (F-122) is presented in Fig.
4. The size distribution is bell-shaped (normal distribution) showing 240 Am as the mean particle diameter. The 90% of population has the size range
between 198 and 307 Am. These results are comparable with the size measurements done by optical
microscope and hence, diameters of the microparticles
calculated before and after swelling are quite reliable.

14

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

Fig. 2. DSC tracings of pure DS (curve a), DS-loaded IPN microspheres containing 10 (curve b), 20 (curve c) and 30% PAA (curve d).

3.6. Microscopic study

Particle size was measured by using optical microscopy and these results are presented in Tables 1
and 2. Size of the particles depends upon the amount
of drug present, % PAA content and extent of crosslinking. Particles are spherical in shape with their
sizes ranging from 160 to 304 Am. The results of %
encapsulation efficiency, % drug loading and mean
particle sizes for different formulations are presented
in Table 1. Particle size of the neat PVA is smaller
than the microspheres containing different amount of
PAA. With an increase in PAA content of the microspheres, size of microspheres increases from 256 to
304 Am for 20% drug containing microspheres (F-132
to F-332). This can be explained on the basis of
hydrodynamic viscosity concept, i.e., as the amount
of PAA in the microspheres increases, interfacial
viscosity of the polymer droplets in the emulsion also

increases because PAA has more water-uptake capacity than PVA, which might hinder the breaking of
dispersed phase into smaller size particles during
emulsification.
For all formulations, with increasing amount of
drug in the microspheres, particle size also increases.
For instance, in case of neat PVA, when the drug
content increased from 5% to 20%, particle size
increased from 160 to 192 Am (F-012 to F-032);
for 10% PAA containing microspheres, particle size
increased from 224 to 256 Am (F-112 to F-132), and
the same trend is also observed for all the other
formulations (see Table 1). This is attributed to the
fact that drug molecules might have occupied the
free volume spaces within the IPN matrix, thereby
hindering the inward shrinkage of the polymer matrix [20]. The 20% drug-loaded and 30% PAAcontaining microspheres (F-332) have the maximum
size of 304 Am.

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

15

groups within the polymeric network ionize and

attract cations into the gel region to replace H+ ions
since the pH of the environmental solution rises above
its pKa value. This effectively increases the concentration of free ions inside the gel. Thus, the ionic
swelling pressure will increase and so does swelling.
Additionally, the gel tends to expand and thereby
minimizes the repulsion between the ionized polycarboxylic groups.
3.7. Encapsulation efficiency

Fig. 3. SEM photograph of microspheres.

Extent of crosslinking has shown an effect on the

particle size (see data in Table 2). For instance, for
microspheres containing 10% PAA and 5% drug, with
increasing crosslinking by GA, i.e., by adding 2.5 to
7.5 ml of GA, the particle size decreased from 240 to
192 Am for formulations F-111 to F-113. This is
attributed to the fact that with an increase in the
amount of GA, the shrinkage of particles might have
occurred leading to the formation of smaller particles
[20,21].
The pKa of PAA is 4.28 [22 24]. When pH is less
than pKa, the H+ ion strength will be high, which will
effectively suppress the ionization of polycarboxylic
acid groups. The gel is neutral and flexibility of the
polymeric chain is rather low. Polycarboxylic acid

Three different concentrations of drug, i.e., 5, 10

and 20 mass% were loaded during crosslinking. The
results of % encapsulation efficiency are also included
in Table 1. These data show an increase with increasing drug loading. In the neat PVA microspheres, as the
drug content increases from 5 to 20 mass%, the
encapsulation efficiency increases from 51% to 58%
(F-012 to F-032). The % encapsulation efficiency also
increases with an increasing amount of PAA in the
microspheres. For instance, to study the effect of PAA
in the microspheres, i.e., for microspheres containing
0%, 10%, 20% and 30% PAA and 5% of DS (i.e., for
formulations F-012, F-112, F-212 and F-312), encapsulation efficiencies are respectively, 51.0%, 71.9%,
80.6% and 86.5%.
For 10% PAA containing PVA microspheres, the
results of effect of extent of crosslinking on size and
entrapment efficiency of the microspheres are presented in Table 2. With increasing crosslinking, the %
encapsulation efficiency has decreased, i.e., for microspheres crosslinked with 2.5, 5 and 7.5 ml of GA (F111, F-112 and F-113), entrapment efficiencies are
respectively, 75.5%, 71.9% and 56.0%. This is be-

Fig. 4. Particle size distribution for formulation containing 10%

PAA, 10% of the drug and 5 ml of GA analyzed by laser light
scattering method.

16

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

Table 2
Results of % encapsulation efficiency and mean size of the
microspheres with different amount of crosslinking agent for
microspheres containing 10% PAA and 5% drug at the stirring
speed of 400 rpm
Formulation Crosslinking %
code
agent
Diclofenac
(GA in ml) sodium
loaded

%
Encapsulation
efficiency
F S.D.

Mean
particle
size
F S.D.

F-111
F-112
F-113

75.5 F 0.5
71.9 F 1.2
56.0 F 0.9

240 F 0.1
224 F 0.8
192 F 1.2

2.5
5
7.5

5
10
20

cause with an increase in crosslinking, microspheres

become more rigid and hence, the free volume space
within the polymer matrix decreases to give reduced
encapsulation efficiency.
3.8. Swelling study
Dynamic equilibrium swelling experiments performed in gastric and intestinal pH conditions for
the microspheres are presented in Tables 3 and 4.
Equilibrium swelling of the microspheres did not
exert any considerable effect in the swelling of microspheres for neat PVA, while for the remaining formulations, equilibrium swelling differed widely. For
all formulations, swelling is more in 7.4 pH media
when compared to swelling in 1.2 pH media; equilibrium swelling is least for pure PVA and, it has not
shown much effect with the pH of the external media.
For all formulations, with increasing amount of DS in
the neat PVA, only a small increase in swelling is
Table 3
Results of equilibrium swelling in different external pH media
Formulation code

F-012
F-022
F-032
F-112
F-122
F-132
F-212
F-222
F-232
F-312
F-322
F-332

Equilibrium swelling ( Q)
HCl (0.1 N)

Buffer (7.4 pH)

58
59.4
60.2
73.1
70.5
70.0
69.5
68.2
65.3
65.4
62.4
61.9

59
59.3
61
118
121
125
140
138
130
169
165
152

observed from 58% to 60% (F-012 to F-032) in 1.2

pH. However, in 7.4 pH media the % increase in
swelling ranges from 59 to 61. As the PAA content
increases from 10% to 30% (F-112 to F-332), swelling
decreases from 73.1% to 61.9% in 1.2 pH media
while it increases from 118% to 152% in 7.4 pH
media. This drastic difference is due to the presence of
COOH groups, which are responsible for increased
hydrophilic nature of the matrix.
Different amounts of crosslinking agent added to
produce microspheres containing 10% PAA and 5%
drug are presented in Table 4. The extent of crosslinking has also shown much difference in equilibrium
swelling, Q, only in 7.4 pH buffer media, whereas for
1.2 pH media, the difference in Q is not considerable.
A swelling of 74.1% is observed for microspheres
containing 2.5 ml of GA (F-111) in 1.2 pH media,
which decreased as the extent of crosslinking increased, i.e., for 5 and 7.5 ml of GA (F-112, F-113)
containing microspheres, swelling is 73.1 and 60.5,
respectively. In 7.4 pH media, swelling decreases
abruptly with an increase in crosslinking, i.e., for
2.5, 5 and 7.5 ml GA containing microspheres (F111, F-112 and F-113), observed swelling is 140%,
118% and 91%, respectively.
Fig. 5 displays the pulsatile behavior with changing pH of the external media for the formulation
containing 10% PAA that was crosslinked with 5 ml
GA (F-101). Swelling increases in 7.4 pH media and
it reaches equilibrium in about 10 min. When the pH
of the media is changed to 1.2, swelling decreases
considerably. Such a behavior was also observed
earlier for pAAm-g-GG microspheres [11].
The results of equilibrium swelling diameter, Dl
normalized to original diameter, D0 are presented in
Table 5. Triplicate measurements gave errors within
2%. Dynamic swelling of the microspheres was de-

Table 4
Results of equilibrium swelling in different pH of the external media
for microspheres containing 10% PAA and 5% drug containing
microspheres with different amount of the crosslinking agent
Formulation code

GA (ml)

Equilibrium swelling ( Q)
HCl (0.1 N)

Buffer (7.4 pH)

F-111
F-112
F-113

2.5
5
7.5

74.1
73.1
60.5

140
118
91

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

17

Fig. 5. Pulsatile behavior plot of normalized diameter vs. swelling

time of the formulation containing 10% PAA and 5 ml of GA
without the drug (F-101) with respect to change in the pH of the
external media (Di is initial diameter).

termined by monitoring the changes in microsphere

diameter, Dt with time using an optical microscope.
Fig. 6 displays the normalized diameter, Dt/D0 (where
D0 is initial diameter) as a function of time for
microparticles containing 10% PAA, without DS that
are crosslinked with different amounts of GA (F-101,
F-102 and F-103). As the amount of GA increases, the
swelling capacity of microspheres decreases considerably. These data are higher in 7.4 pH media than
observed in 1.2 pH media. As the amount of GA in
the microspheres increases from 2.5 to 7.5 ml, equilibrium normalized diameter decreases from 1.5 to
1.24 in 7.4 pH and from 1.23 to 1.18 in 1.2 pH media
(see Table 5).
Dimensional changes of the microspheres due to
swelling (i.e., volume change, DVt with time from the
initial volume, V0) have been measured and used to
compute the diffusion coefficient, Dv of the drug
Table 5
Transport data of microspheres containing 10% PAA without drug
and crosslinked with different amounts of GA in 7.4 and 1.2 pH
conditions
pH

GA used
(in ml)

Equilibrium
normalized
diameter (Dl/Do)

Dv  105 (cm2/s)

7.4

2.5
5.0
7.5
2.5
5.0
7.5

1.50
1.27
1.24
1.23
1.21
1.18

3.95
1.94
1.39
1.50
1.24
0.27

1.2

Fig. 6. Plot of normalized diameter vs. swelling time for

formulations, F-101 (.), F-102 (E) and F-103 (n) at 7.4 (A) and
1.2 (B) pH media.

containing aqueous media using the theory proposed

by Harogoppad and Aminabhavi [25].
3
2 




1=2
l
4 DV
V0
DVt
Dv
4
5
t 1=2
2

D0
V0
p

V0 D0
Dv 1:773  Slope
4DVl

2
3

Here, DVl is the change in volume at equilibrium, Vl

is volume at equilibrium swelling. Eq. (3) is used to
calculate Dv from the slope of the initial linear plots of
DVt/Vo vs. t1/2. These data are also included in Table 5.
Diffusion coefficients show systematic variations
with the amount of GA added in the matrix. For
instance, values of Dv decrease with increasing
amount of GA in the microspheres containing 20%
PAA and 5% of DS. This explains the conventional
wisdom at higher crosslinking, free volume of the

18

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

To understand the drug release from the DS-loaded

sequential IPN microspheres of PVA and PAA, in
vitro release experiments were carried out in gastric
and intestinal pH conditions. These results exhibit a
phenomenal effect of the matrix crosslinking on drug

release characteristics for all the formulations. Fig. 7

displays the release profiles of microspheres crosslinked with different amounts of GA containing 10%
PAA and 5% DS (F-111, F-112 and F-113) in 7.4 pH
media with respect to time. The cumulative % release
is higher in case of microspheres crosslinked with 2.5
ml GA (F-111), the least % release is observed with
microspheres crosslinked with 7.5 ml GA (F-113),
while the intermediary values are observed for microspheres crosslinked with 5 ml of GA (F-112). This is
due to a decrease in swelling as the amount of GA
increases in microspheres. The % cumulative release
data were obtained in triplicate (with < 3% deviations
in all cases).
Figs. 8 and 9 display the cumulative % release data
of PVA microspheres for different amounts of PAA
containing 10% DS and 5 ml of GA (F-022, F-122, F222 and F-322) respectively, at 1.2 and 7.4 pH media
with respect to time. A pronounced difference is
observed in the release data at 1.2 and 7.4 pH, which
is attributed to the presence of COOH groups that are
responsible for higher swelling in higher pH media.
Increased cumulative release is observed as the amount
of PAA in the microsphere increases. This is due to the
increased COOH groups with increasing PAA content
thereby inducing higher water-uptake capacity of the
microspheres; this consequently increases matrix
swelling. In both the release media of 1.2 and 7.4 pH,
the neat PVA microspheres showed the least % cumulative release, while 10% and 20% PAA-containing
PVA microspheres showed the intermediate % cumu-

Fig. 8. In vitro % cumulative release in 1.2 pH vs. time for

formulations, F-022 (.), F-122 (E), F-222 (n) and F-322 (x).

Fig. 9. In vitro % cumulative release in 7.4 pH vs. time for

formulations, F-022 (.), F-122 (E), F-222 (n) and F-322 (x).

Fig. 7. In vitro % cumulative release vs. time for formulations, F-111

(.), F-112 (E) and F-113 (n).

matrix will be less thereby, hindering the easy transport of molecules through the matrix. It may be noted
that Dv values in 7.4 pH media are higher than those
observed in 1.2 pH media indicating that in acidic
media more of solvent molecules transport than in the
basic media.
3.9. In vitro release study

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

lative release and 30% PAA-containing PVA microspheres showed the highest % cumulative release. The
data points presented in Figs. 6 9 represent averages of
triplicate measurements obtained within 3% standard
deviations.

4. Conclusions
The pH-sensitive PVA-PAA sequential IPN microspheres were prepared and used in the controlled
release of DS. Due to the presence of ionizable
carboxylic functional groups, the microgels formed
were sensitive to pH as well as ion strength of the
external media. The pH-sensitivity of microgels was
evaluated by monitoring their dimensional changes
with time using light microscopy. The release of DS
from microgels was dependent upon the pH of the
medium, extent of crosslinking and the amount of
drug loading.

Acknowledgements
The authors thank the University Grants Commission (Grant No. F. 1-41/2001 (CPP-II)) for awarding a
major grant to establish Center of Excellence in
Polymer Science on the campus of Karnatak University, Dharwad. Authors are also thankful to Dr. A.R.
Kulkarni for his assistance in this project.

References
[1] K. Abu-Izza, L. Tambrallo, L.D. Robert, In vivo evaluation of
zidovudine (AZT)-loaded ethylcellulose microspheres after
oral administration in beagle dogs, J. Pharm. Sci. 86 (1997)
554 559.
[2] I. Ghebre-Sellassie, Multiparticulate Oral Drug Delivery, Marcel Dekker, New York, 1994, p. 480.
[3] R. Bodmeier, H. Chen, P. Tyle, P. Jarosz, Pseudoephedrine
hydrochloride microspheres formulated into an oral suspension dosage form, J. Control. Release 15 (1990) 65 77.
[4] N. Nyamweya, K.A. Mehta, S.W. Hoag, Characterization of
the interactions between polymethylacrylate-based aqueous
polymeric dispersions and aluminum lakes, J. Pharm. Sci.
90 (2001) 1937 1947.
[5] K.J. Brodbeck, S. Pushpala, A.J. McHugh, Sustained release
of human growth hormone from PLGA solution depots,
Pharm. Res. 16 (1999) 1825 1829.

19

[6] S.W. Kim, in: N. Ogata, S.W. Kim, J. Feijen, T. Okano (Eds.),
Biomed. Eng. Drug Delivery Systems, Springer-Verlag,
Tokyo, 1996.
[7] A.S. Hoffman, Intelligent polymers in medicine and biotechnology, Macromol. Symp. 98 (1995) 645 664.
[8] R. Bettini, P. Colombo, N.A. Peppas, Solubility effects on
drug transport through pH-sensitive swelling controlled release systems. Transport of theophylline and metoclopramide monohydrochloride, J. Control. Release 37 (1995)
105 111.
[9] C. Ramakisson-Ganorkar, F. Liu, M. Baudys, S.W. Kim, Modulating insulin-release profile from pH/thermosensitive polymeric beads through polymer molecular weight, J. Control.
Release 59 (1999) 287 298.
[10] X. Qu, A. Wirsen, A.C. Albertsson, Novel pH-sensitive chitosan hydrogels: swelling behavior and states of water, Polymer 41 (2000) 4589 4598.
[11] K.S. Soppimath, A.R. Kulkarni, T.M. Aminabhavi, Chemically modified polyacrylamide-g-guar gum based cross-linked
anionic microgels as pH-sensitive drug delivery systems: preparation and characterization, J. Control. Release 75 (2001)
331 345.
[12] Y.M. Lee, J.K. Shim, Plasma surface graft of acrylic acid onto
a porous poly(vinylidene fluoride) membrane and its riboflavin permeation, J. Appl. Polym. Sci. 61 (1996) 1245 1250.
[13] R. Aliev, P. Garca, G. Burillo, Graft copolymerization of
acrylic acid onto polycabonate by the preirradiation method,
Radiat. Phys. Chem. 58 (2000) 299 304.
[14] A.R. Kulkarni, K.S. Soppimath, T.M. Aminabhavi, Controlled
release of diclofenac sodium from sodium alginate beads
crosslinked with glutaraldehyde, Pharm. Acta Helv. 74
(1999) 29 36.
[15] K. Chandermun, C.M. Danprox, T. Govender, The effect of
selected formulation and process variables on the release characteristics of pellets produced by extrusion-spheronisation,
Proc. Int. Symp. Control. Release Bioact. Mater. 25 (1998)
942 943.
[16] A.R. Kulkarni, K.S. Soppimath, T.M. Aminabhavi, Ureaformaldehyde nanocapsules for the controlled release of diclofenac sodium, J. Microencapsul. 14 (2000) 449 458.
[17] S.G. Kumbar, A.R. Kulkarni, T.M. Aminabhavi, Crosslinked
chitosan microspheres for encapsulation of diclofenac sodium:
effect of crosslinking agent, J. Microencapsul. 19 (2002)
173 180.
[18] C.K. Yeom, R.Y.M. Huang, Pervaporation separation of gaseous mixtures using crosslinked poly(vinyl alcohol) and amic
acid, Angew. Makromol. Chem. 184 (1991) 27 40.
[19] K.S. Soppimath, A.R. Kulkarni, T.M. Aminabhavi, Controlled
release of antihypertensive drug from the interpenetrating network poly(vinyl alcohol) guar gum hydrogel microspheres,
J. Biomater. Sci., Polym. Ed. 11 (2000) 27 43.
[20] K.S. Soppimath, A.R. Kulkarni, T.M. Aminabhavi, Water
transport and drug release study of crosslinked guar gum
grafted polyacrylamide hydrogel microspheres for the controlled release applications, Eur. J. Pharm. Biopharm. 53
(2002) 87 98.
[21] R.C. Korsmeyer, N.A. Peppas, Effect of morphology of hy-

20

M.D. Kurkuri, T.M. Aminabhavi / Journal of Controlled Release 96 (2004) 920

drophilic polymeric matrices on the diffusion and release of

water soluble drugs, J. Membr. Sci. 9 (1981) 211 227.
[22] K. Kajiwara, S.B. Rose-Murphy, Synthetic gels on the move,
Nature 355 (1992) 208 209.
[23] T. Shiga, K. Fukumori, Y. Hiorse, A. Okada, T. Kurauchi,
Pulsed NMR study of the structure of poly(vinyl alcohol)
poly(sodium acrylate) composite hydrogel, J. Polym. Sci., B,
Polym. Phys. 32 (1994) 85 90.

[24] J. Fei, Z. Zhang, L. Zhong, L. Gu, PVA/PAA thermo-induced

hydrogel fiber: preparation and pH-sensitive behavior in electrolyte solution, J. Appl. Polym. Sci. 85 (2002) 2423 2430.
[25] S.B. Harogoppad, T.M. Aminabhavi, Diffusion and sorption
of organic liquids through polymer membranes: VIII. Elastomers versus monocyclic aromatic liquids, J. Appl. Polym. Sci.
46 (1992) 725 732.