Drug Delivery

ISSN: 1071-7544 (Print) 1521-0464 (Online) Journal homepage: https://www.tandfonline.com/loi/idrd20

Silica-Polyethylene Glycol Matrix Synthesis by

Sol-Gel Method and Evaluation for Diclofenac
Diethyloammonium Release

Magdalena Prokopowicz

To cite this article: Magdalena Prokopowicz (2007) Silica-Polyethylene Glycol Matrix Synthesis
by Sol-Gel Method and Evaluation for Diclofenac Diethyloammonium Release, Drug Delivery, 14:3,
129-138, DOI: 10.1080/10717540600812653

To link to this article: https://doi.org/10.1080/10717540600812653

Published online: 10 Oct 2008.

Submit your article to this journal

Article views: 1436

View related articles

Citing articles: 11 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=idrd20
Drug Delivery, 14:129–138, 2007
Copyright

c Informa Healthcare
ISSN: 1071-7544 print / 1521-0464 online
DOI: 10.1080/10717540600812653
Silica-Polyethylene Glycol Matrix Synthesis
by Sol-Gel Method and Evaluation
for Diclofenac Diethyloammonium Release
Magdalena Prokopowicz
Division of Physical Chemistry with Instrumental Analysis Laboratory,
Medical Academy of Gdańsk, Gdańsk, Poland
Modified silica-polyethylene glycol xerogels were prepared by
the sol-gel method to explore the possibilities of using these
polymers as drug delivery systems. The synthesis was per-
formed at room temperature and under atmospheric pressure us-
ing tetraethylorthosilicate (TEOS) as a precursor, low-molecular
polyethylene-glycol (600) as a modifier, and acetic acid as a cata-
lyst. The composition in a mole ratio of the initial sols corresponds
to TEOS:H2 O:EtOH:CH3 COOH:PEG = 1:4:6:0.005:0.147. Diclo-
fenac diethyloammonium was used as a model drug and encapsu-
lated in predoping sol-gel process. After 5 days of gelation time of
matrices at room temperature two different forms of xerogels were
obtained: monolithic form of pellet and cracked, irregular-shaped
of particles. The rate of release from the both forms of xerogels
was controlled by the rate of diffusion of the drug through the
pores. The dissolution testing for the loaded irregular-shaped xe-
rogels showed an initial burst release followed by sustained release.
The degradation of the PEG/silica xerogels followed a zero-order
kinetics.
Keywords Diclofenac Control Release, Silica-Polyethylene Glycol
Xerogels, Silica Polymers, Sol-Gel Methods
An increased interest has been observed in silica and silica-
organic biomaterials synthesized by the sol-gel method and their
potential application to deliver a biologically active substance
precisely to a targeted site or to be used as an implantable drug
delivery system (Ahola et al. 2000; Ahola et al. 2001; Kortesuto
et al. 2001; Byrne and Deasy 2002; Prokopowicz, L
ukasiak,
and Przyjazny 2004). The principal advantage of synthesis of
biomaterials using the sol-gel method is that the method is un-
complicated, it takes place at room temperature and it does not
require the use of pharmaceutically unacceptable solvents. The
most commonly used precursor in the synthesis of such gels is
Received 19 January 2006; accepted 8 May 2006.
Address correspondence to M. Prokopowicz, Medical Academy of
Gdańsk, Division of Physical Chemistry, Hallera 107, 80-416 Gdańsk,
Poland. Fax: +48-58-3493206. E-mail: mprokop@biology.pl
129
tetraethylorthosilicate (TEOS), and the solvents are water and
ethanol. The sol-gel process is initiated by adding water to the
alcoholic solution of TEOS. Following hydrolysis and polycon-
densation reactions, a two-phase system is obtained as a result of
gel formation. Hydrolysis and condensation of silicon alkoxide
are usually enhanced by acid or base catalysis (Livage 1992).
Hydrolysis is favored by acid catalysis so that Si(OH)4 species
can be obtained when excess water is added. Condensation pro-
ceeds faster in the presence of basic catalysts.
The obtained system consists of inorganic porous siloxane
skeleton and the solvent entrapped within it. Amorphous bio-
materials obtained in this way are characterized by high purity
and homogeneity (Lin and Brown 1997; Livage 1997), in vivo
biocompatibility (Wilson et al. 1981; Palumbo 1997; Kortesuo
et al. 2000), and slow degradation (Kortesuo et al. 2000; Radin
et al. 2002). A biologically active substance can be introduced
into a gel either at the stage of sol formation or after the for-
mation of a porous xerogel. In the former case (predoping), the
molecules of a substance are entrapped within the porous sil-
ica gel network as a result of polycondensation (Lin and Brown
1997; Livage 1997; Dunn et al. 1998). In the latter case, called
impregnation (postdoping), the gel matrix is saturated with a
substance after being immersed in its solution. The substance
molecules penetrate into the pores in a diffusion process de-
pending on the pore diameter and the viscosity of the solution
(Lin and Brown 1997; Livage 1997).
The process of making silica gel matrices can be modified by
varying physicochemical parameters of gelation, such as tem-
perature, kind of reactants used and their ratio, pH of the solu-
tion, and the kind of catalyst used (Brinker 1996; Higginbotham
et al. 2003; Leonard et al. 2003). These modifications have an
indirect effect on physicochemical properties of the silica poly-
mer obtained, including specific surface area, kind, size, and
distribution of the pores; degree of cross-linking of the gel; re-
fractive index; and hardness. An important factor influencing
the structure of silica xerogels is pH of the solution. With an
acidic catalyst, rapidly proceeding hydrolysis causes the forma-
tion of linear polymers, which in consequence creates a solid
130 M. PROKOPOWICZ
material with less porosity but of improved mechanical strength
compared with products of basic catalysis. In contrast, basic
catalysis favors the formation of an agglomerate consisting of
branched colloidal particles, which yields a porous material with
high adsorption capacity (Lin and Brown 1997; Livage 1997).
However, to obtain repeatable and reproducible batches of
the material, the steps and conditions of gel synthesis should
be closely followed (Brinker 1996; Higginbotham et al. 2003;
Leonard et al. 2003). The last step affecting the final form of
the gels is the removal of solvent residues, especially water. A
proper selection of the drying technique is of utmost importance,
because the encapsulated biologically active substance is usually
thermosensitive and traditional drying at an elevated temperature
is impossible.
The most common technique of drying the gels with an en-
capsulated substance is by placing them in a desiccator at room
temperature, in the vacuum oven, or by using a supercritical
fluid in an autoclave (Rao and Chung 1991; Venkateswara et al.
1999; Morris et al. 2001; Prokopowicz, L
ukasiak, and Przyjazny
2005). Depending on the sol-gel synthesis method and the dry-
ing technique used, different forms of the gel can be obtained,
such as powders, fibers, layers, and thin films and monoliths
and porous membranes. The encapsulated molecules of a bi-
ologically active compound can be released from the gels by
diffusion, which is affected by the pore size and their inter-
connections, the shape of the matrix, as well as the size of the
molecule being released (Lin and Brown 1997).
An additional advantage of the sol-gel method is the possi-
bility of synthesis by using a combined sol-gel processing in tar-
geted formation of organosiloxane hybrids of specific mechan-
ical and chemical properties (Guo, Hyeon-Lee, and Beaucage
1997; Netz et al. 2001; Ahola et al. 2001; Higginbotham et al.
2003). In this article, the traditional sol-gel synthesis developed
by Brinker et al. (1984) was modified by the addition of a low-
molecular organic modifier of a hydrophilic nature (polyethy-
lene glycol, PEG, 600 g/mol) and by using ambient conditions
to dry the PEG/siloxane hybrids. The addition of polyethylene
glycol as a modifier to pure silica has an effect on the formation
of silica particles of the colloidal size during polycondensation
through the formation both of covalent bonds with silicon alkox-
ide and hydrogen bonds with residual silanol groups in the silica
lattice. It is also likely to have an effect on both the number and
average size of pores of the obtained xerogel.
Diclofenac diethyloammonium, a water-soluble nons-
teroidal anti-inflammatory drug (NSAID) with a strong anti-
inflammatory and pain killing activity, was selected for encap-
sulation. It is most often used to reduce pain, inflammation,
and stiffness caused by many conditions, such as osteoarthritis
or rheumatoid arthritis, as well as a pain killer in some types of
cancer (Oberle et al. 1984; Liu et al. 1995). As a result of its short
half-life, the daily dose of the drug for an adult with rheuma-
toid arthritis is 75–100 mg/24 hr. However, long-term taking
of diclofenac in traditional forms (oral, injections) may result
in toxicity to the human tissues (Witham 1991; Gómez-Lechón
et al. 2003; Schwaiger et al. 2004). Among all NSAID drugs,
diclofenac is the medication that most often damages the liver
(Gómez-Lechón et al. 2003). It also can destroy hepatocytes as
well as induce their apoptosis (Schwaiger et al. 2004).
The modifications in diclofenac release from solid forms of
the medication used thus far involve coating the drug with macro-
molecular compounds, e.g., polymethacrylate, forming insol-
uble layers delaying drug release (Kramar, Turk, and Vrečer
2003), or incorporation into a carrier (Boinpally et al. 2003)
through the formation of capsules filled with pellets or granules
(Fang et al. 1999). This method increases bioavailability of the
medication and reduces side effects and local irritation of the
mucosa.
The use of implants with an encapsulated pain killer or com-
posite chemotherapeutic and pain killer also is becoming ap-
preciated in palliative care of cancer patients not tolerating oral
drugs and when the life of the patient cannot be significantly
extended, but its quality is an issue. In this application the dose
control is not that important, but it is essential that the patient
could take drugs at home, in which case the quality of life could
be estimated by, for example, the pain killing effect. The prepa-
ration and utilization of the nonmodified acid-catalyzed silica
gel, base-catalyzed gel, and noncatalyzed gel for a long-term
release of a chemotherapeutic—doxorubicin—was described in
a previous work (Prokopowicz et al. 2004).
The aim of this study was to prepare acid-catalyzed silica gel
matrix modified with low-molecular PEG (600) and evaluate the
suitability of this matrix as a carrier material for the controlled
release of the pain killer diclofenac. First, the polymer structure
by FTIR and SEM was investigated. Then the effects of the
shape of obtained PEG/xerogel on the encapsulation efficiency,
the release study of drug, and the stability of matrixes were
determined.
MATERIALS AND METHODS
Reagents
All reagents and the drug were obtained from Sigma Chem-
ical Company and used without further purification: TEOS,
M = 208.33 g/mol; polyethylene glycol (PEG), M = 600 g/mol;
ethanol 98%(v/v); acetic acid 0.1 M; deionized water, and
diclofenac diethyloammonium.
Simulated body fluid (SBF) had the following composi-
tion: NaCl (136.8 mM), NaHCO3 (4.2 mM), KCl (3.0 mM),
K2 HPO4 · 3H2 O (1.0 mM), MgCl2 · 6 H2 O (1.5 mM), CaCl2 · 2
H2 O (2.5 mM) and Na2 SO4 (0.5 mM). It was buffered at pH
7.4 with phosphate buffer (50 mM). A Millipore Milli Q-plus
system was used to purify water for the preparation of buffers
and reagent solutions (resistance of water 18 M
).
Spectrophotometric Measurements
The quantitative measurements of diclofenac diethylammo-
nium were performed using an HP 8452A Diode-Array UV/VIS
 SILICA-POLYETHYLENE GLYCOL FOR DICLOFENAC RELEASE 131

spectrophotometer connected to an IBM Pentium 100 computer.
The analytical wavelength λ = 276 nm corresponding to the ab-
sorbance maximum was selected for the determination of the
drug.
Preparation of Standard Solutions of Diclofenac
Diclofenac diethyloammonium in amount of 0.05 g ± 0,001 g
was dissolved in 100 ml of a deionized water solution at room
temperature. (The saturation of drug in the water was in the con-
centration >0.05 g/mL indicating the sink condition limits for
the dissolution studies). To determine spectrophotometric mea-
surements of diclofenac diethylammonium, standard solutions
of the drug were then prepared from the stocks solution by dilu-
tion with the water. We found that the dependence of absorbance
on the concentration of standard solutions of diclofenac ranging
from 0.003 mg/mL to 0.15 mg/mL adhered to the Lambert-Beer
law, and nonlinearity was observed for the concentration above
0.15 mg/mL. The calibration curves were prepared and used for
the determination of the amounts of diclofenac released.
Synthesis of Diclofenac-Loaded Xerogels
Diclofenac-loaded xerogels were prepared through the sol-
gel process, by hydrolysis, and polycondensation of reagents:
TEOS (Si(OC2 H5 )4 ) diluted in water, ethanol, and polyethylene
glycol (H(OCH2 CH2 )n OH) in the presence of acetic acid catalyst
(pH = 4) using the following chemical reactions:
• Hydrolysis:
≡Si OC2 H5 + H2 O
≡ Si OH + C2 H5 OH
• Condensation:
≡Si OC2 H5 + HO Si≡
≡Si O Si≡
+C2 H5 OH
≡Si OH + HO Si≡
≡Si O Si≡ + H2 O
≡Si OC2 H5 + H(OCH2 CH2 )n OH


≡Si O CH2 (CH2 O CH2 )n CH2 O Si≡
+ C2 H5 OH
All gels were synthesized at room temperature and under
atmospheric pressure using the following procedure. Ethanol
(8.11 g) was added dropwise to polyethylene beakers (50 mL)
containing PEG (2.93 g) and the solution was stirred for 15
min using a magnetic stirrer. Next, TEOS (6.90 g) and the
solution containing redistilled water (2.4 g) with the acetic
acid catalyst (0.30 g) were added dropwise, and the mix-
ture was stirred for further 15 min. The pH of the acid
solutions was 4. Thus obtained mixture had a mole ra-
tio of TEOS:H2 O:EtOH:CH3 COOH:PEG = 1:4:6:0.005:0.147.
to continue polycondensation, gelation, and aging (during this
time the strength of the gel thereby increases (Lin and Brown
1997)). Next, the moulds with the gels were opened and kept
for further aging and drying under ambient conditions (in the
drying process the solvents [water and ethanol] are removed
from the interconnected pore network [Lien and Brown 1997]).
Following conditioning for a prescribed time, so-called xero-
gels were obtained. The end of conditioning of the gels with
encapsulated diclofenac was considered to be the moment when
the solids with a mass constant within the limits of instrumental
error (±0.1 mg) were obtained.
Next, the diclofenac-loaded xerogels were rinsed with deion-
ized water (for 5 sec) to remove unreacted reagents and kept in
a desiccator (over P2 O5 ) prior to studies. Prior to studies the
irregular-shaped xerogels and pellets were weighed and the di-
mensions of each pellet were measured by a micrometer screw
gauge.
Three independent syntheses were carried out using the pro-
cedure described above and the average mass of diclofenac-
loaded xerogels obtained form the mould with sol was evalu-
ated for 10 samples. A schematic diagram of preparation of the
diclofenac-loaded PEG-silica gel is shown in Figure 1.
ETHANOL + (PEG 600)
(room temperature)
STIRRING (15 min.)
TEOS + DE-IONIZED
WATER (+CH3COOH)
STIRRING (15 min.)
WATER SOLUTION OF
DICLOFENAC
DIETHYLOAMMONIUM
SONICATION
(8-10 min.)
CASTING
(gelating, aging)
(room temperature)
DRYING
(ambient conditions)

Following stirring, the mixtures were pipetted into round,

polystyrene moulds (2 mL/mould) and 100 µL of variable con-
centrations of a standard diclofenac solution (0.008 g/mL, 0.004 STORAGE
g/mL, respectively) were added dropwise to the mould. (DESICCATOR)
Next, the moulds with mixtures were tightly closed and soni-
cated in a cold water bath until they were clear and homogeneous FIG. 1. Schematic diagrams of the diclofenac-loaded PEG/silica gel
(8–10 min) and kept for a prescribed time at room temperature preparation.
132 M. PROKOPOWICZ
Synthesis of Pure Gel without Diclofenac
Control samples (pure PEG/silica xerogels) were examined at
the same time as the investigated samples to check the behavior
of pure gels during incubation with the buffer under the same
conditions. To this end, in the step of formulation of the gels
described above the sol was mixed with 100 µL of water without
the drug and the rest of the procedure was exactly the same.
Characteristics of Diclofenac-Loaded Xerogels
The time of gel preparation was measured from the moment
of mixing all the reagents until obtaining mature gels of vitre-
ous appearance dried to constant mass. The specific surface area
was measured using the BET method based on nitrogen gas ad-
sorption (Micromeritics ASAP 2000). Before the measurement,
samples were vacuum dried for 24 hr at 25◦ C. The bulk den-
sity was measured from the known volumes and weights of the
xerogels.
The morphology of the biomaterials was observed by scan-
ning electron microscopy (ESEM, Philips) at acceleration volt-
age of 20 kV with variable vacuum in the atmosphere of water
vapor. Samples for SEM were prepared by mounting on silicon
stubs. The FTIR spectra of samples of PEG/silica xerogels were
taken between 650 and 4000 cm−1 using a Jasco model 410
FTIR in transmission mode and KBr as a background material.
The KBr pellet of each sample was prepared using 100 mg of
KBr and 1 mg of powdered sample of pure silica xerogel (with-
out PEG and drug) and 1.5 mg of PEG/silica xerogels with drug.
The resolution of the FTIR instrument was 4 cm−1 .
Efficiency Encapsulation of Diclofenac
To investigate the total amount and efficiency of encapsu-
lation of diclofenac during sol gelation, some of the weighed
diclofenac-loaded PEG/silica gels (3 independent samples) were
crushed and ground to a powder. The powder was added to
100 mL of the phosphate buffer (pH 10) followed by 30-min
sonication. The total amount of free drug was determined spec-
trophotometrically after separation of particles from the aqueous
medium by ultracentrifugation and after appropriate dilutions.
The actual loading was expressed as the average (three measure-
ments) percentage of the diclofenac released from the weighed
gel (%w/w). The actual loading efficiency was calculated based
on the percent ratio of the amount of drug incorporated into the
gels to a theoretical (initial) amount used. The drug-loading effi-
cacy equal to 100% indicates that the entire drug was entrapped
in the PEG/silica gels.
Release Study
The release study of the drug from diclofenac-loaded
PEG/silica gel was performed in an incubator (Tropicool clas-
sic, Poland) at 37 + 1.5◦ C. The medium into which the drug
was released was 20 mL of SBF (pH 7.4). The release medium
was stirred magnetically. At different time intervals, 1 mL of
solution was taken for the measurement of the amount of re-
leased diclofenac and replaced by a fresh 1 mL aliquot of SBF
to maintain constant volume and to ensure sink conditions. The
quantitative determination of the amount of diclofenac released
from the matrix was carried out on the basis of precalibration
of the spectrometer using standard solutions of the drug. The
release studies of diclofenac were carried out three times for
3 hr.
Stability Study
The stability study was performed for the PEG/silica xerogels
with the drug and without it. The stability study was determined
by measuring of the dissolved Si(OH)4 as a molybdenum blue
complex by UV/VIS spectroscopy at 814 nm (Koch and Koch-
Dedic 1974). The procedure used was the same as the one de-
scribed in the previous section except that after each collection
of a 10-mL sample, a fresh 10-mL aliquot of SBF was added to
the medium. The degree of degradation of the gel matrix was
expressed as the average (3 measurements) percentage of the sil-
ica acid released from the weighed gel (%w/w). The degradation
studies was monitored over 60 days. Precalibration curves were
used for the determination of the amounts of dissolved Si(OH)4 .
Statistical Analysis
The data obtained from encapsulation efficiency, release rate
determination studies of diclofenac from PEG/silica gels, and
degradation rate of PEG/silica gels without the drug and with
the drug were analyzed statistically by Student’s t-tests using
version 4.2 of Matlab program (Math Works).
RESULTS AND DISCUSSION
As a result of acid-catalyzed (CH3 COOH, pH = 4) sol-gel
synthesis modified by the addition of low-molecular PEG (600)
and drying under ambient conditions, two forms of diclofenac-
loaded gel matrices were obtained: cracked, irregular-shaped,
and transparent forms of particles and monolithic, transparent
pellets. The time of gelation, aging, and drying at room temper-
ature of the gels being equal 5 days.
The average mass of diclofenac-loaded PEG/xerogel ob-
tained from the one mould with sol was 520 ± 12 mg ( p < 0.05).
The photographs of drug-loaded PEG/silica pellet and irregular-
shaped particles are shown in Figure 2.
Using acid-catalyzed solutions and as the modifier low-
molecular PEG in the synthesis step results in the formation of
xerogels with the value of bulk density about 1.5 ± 0.05 g/cm3
( p < 0.05) and a specific surface area 420 ± 7 m2 /g ( p < 0.05).
Furthermore, the adsorption-desorption isotherms of PEG/silica
xerogels were of type 1 form characteristic of a microporous
solid, defined as a material with pores smaller than or equal to
1.5 nm (Brinker and Scherer 1990; Lowell and Shields 1991).
In this work all PEG/silica xerogels were examined by SEM and
the results were found to be very similar. The photographs of se-
lected matrices at different magnification are shown in Figure 3.
 SILICA-POLYETHYLENE GLYCOL FOR DICLOFENAC RELEASE
FIG. 2. Photo of PEG/silica xerogel in form of (a) pellet and (b) irregular-
shaped.
The SEM pictures reveal (Figure 3a) that all the synthesized
xerogels have an external protective coating with wall surfaces
smooth and compacted. However, high magnification SEM im-
ages (Figure 3b) showed the presence of micropores.
The FTIR spectra of selected, obtained samples are shown
in Figure 4. The inspection of the spectra reveals no visible dif-
ferences between characteristic bands in the spectrum of pure
silica matrix (no drug and PEG added) (Figure 4a) and the spec-
trum of the PEG/silica containing the drug (Figure 4b). The in-
FIG. 3. SEM pictures of PEG/silica xerogel: (a) 1 mm and (b) 20 µm.
133
frared absorption bands observed between 3800 and 3000 cm−1
and at 1620 cm−1 correspond to stretching and bending vibra-
tions of the OH group coming, in case of xerogels, from free
or adsorbed water. Drying xerogels at 35◦ C resulted in a low-
ered intensity of the bands (spectrum not shown). The bands
at 2800–3000 cm−1 are characteristic of the C H stretching
vibrations coming from the CH2 and CH3 groups present
in pure precursors (TEOS, PEG) (spectra not shown) and in
xerogels. However, these bands are the most characteristic for
xerogels modified with PEG. The presence of bands at 12500–
1000 cm−1 and 800 cm−1 is characteristic of symmetric and
asymmetric vibrations of the Si O or O Si O groups, while
the band at 950 cm−1 indicates the presence of the Si O H
group. On the other hand, the bands at 1160–1060 cm−1 are at-
tributable to stretching vibrations of the C O and C C groups
coming from PEG.
In summary, the spectra of all PEG/xerogels were typical of
acid-catalyzed, well-hydrolyzed silica xerogels with a high con-
centration of terminal groups (silanols) (Brinker and Scherer
1990). An additional observation was that the spectrum of gel
matrix containing diclofenac did not contain characteristic bands
attributed to the pure drug (spectra of diclofenac diethyloammo-
nium is shown on the Figure 5). One possible explanation is too
small an amount of the drug encapsulated in the matrices, and/or
the possibility that diclofenac can react with the reagents and un-
dergo cross-linking during the formation of the polymeric lattice
of the matrix, and/or the chance that the microporous structure
of the gel shields the drug, thus obstructing its identification by
IR. The UV/VIS spectra of the drug released from the matrices
do not reveal any statistically significant differences in shifts
of peaks attributed to the drug, which could possibly indicate
physicochemical changes in the drug structure.
In addition, preliminary experiments were performed using
C18 reverse phase HPLC with variable wavelength UV detection
aimed at determining the purity of solution containing diclofenac
released from the silica gel matrix. The results obtained revealed
just one peak, attributed to the drug. This confirmed our assump-
tions that the sol/gel synthesis yields gel matrices, which only
encapsulate diclofenac without reactions that could possibly re-
sult in the formation of oligomers of drug or other species of
diclofenac.
The total amount of diclofenac loading and loading efficiency
are compared in Table 1. The loading efficiency of diclofenac
TABLE 1
Actual loading and loading efficiency of diclofenac in
acid-catalyzed diclofenac loaded PEG/silica xerogels
Mass of diclofenac Theoretical Actual loading Loading
added to mould loading per per 500 mg efficiency
with sol (mg) 520 mg (%w/w) (%w/w) (%)
0.4 0.08 0.08 ± 0.01 100 ± 12.5
0.8 0.15 0.16 ± 0.02 107 ± 13.3
(n = 3, mean ± S.D., p < 0.05).
134 M. PROKOPOWICZ

100

Pure silica
80 xerogel

PEG/silica
60
xerogel
with
%T diclofenac

40

20

0
4000 3600 3200 2800 2400 2000 1800 1600 1400 1200 1000 800 700
Wavenumber[cm-1]

FIG. 4. FTIR spectra of silica xerogel without PEG 1 and diclofenac and PEG/silica xerogel with diclofenac.

encapsulation for all PEG/silica gels was close to 100% and
was not affected by the matrix:drug ratio ( p < 0.05). The mean
value of total drug load in the masses of xerogels (520 ± 12 mg)
was 0.08 ± 0.01 [%w/w] and 0.16 ± 0.02 [%w/w], respectively.
We concluded that the acid-catalyzed sol/gel predoping method
enables a complete encapsulation of the drug in the gel without
any measurable losses. The uncertainty of the final result is a sum
of many composites, including random errors associated with
the analytical procedure used (preparation of standard solutions
of drug, dilution), purity of the reagents used, and the errors
in weighing and analyzing samples on an analytical balance as
well as the instrumental errors.
The irregular-shaped diclofenac-loaded PEG/xerogels with
diameters between 1 and 3 mm, and the monolithic pellets with
a diameter 10 ± 0.5 mm and thickness 1.5 ± 0.5 mm were used
for release and stability experiments. Figure 6 illustrate a com-
parison of the rate of release of diclofenac from the irregular-
shaped of matrices with the rate of release of drug from pellets
containing the different amount of diclofenac as the dependence
of the cumulative mean percent of drug released on time. As can
be seen, the drug release rates from PEG/silica xerogels were
affected by the shape of matrices. By examining the release pro-
files from irregular-shaped PEG/silica xerogels in the beginning
period up to about 10 min, we can see the burst release in which

100

80

60

%T

40

20

0
4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400
Wavenumber[cm-1]

FIG. 5. FTIR spectra of diclofenac diethyloammonium.

 SILICA-POLYETHYLENE GLYCOL FOR DICLOFENAC RELEASE 135

pellets 2,5

percent of drug released

(diameter= 10 ± 0.5 mm, thickness = 1.5 ± 0.5 mm) (a)

log of cumulative mean

2
irregular particles
70 (1 mm< diameter< 3mm)
1,5

cumulative mean percent of drug

1
60
0,5
50 0
0 50 100 150
released

40 time of released [min]

30

cumulative mean percent

80
(b)

of drug released
20 60

40
10
20
0
0 50 100 150 0
0 5 10 15
time of released [min] square root of time released

FIG. 6. Cumulative mean percent of diclofenac release from PEG/silica xerogels as a function of total time of incubation in SBF buffer solution. Dates were
given as mean ± S.D. for n = 3. (a) First-order plot of diclofenac released from PEG/silica xerogels. (b) Higushi plot of diclofenac released from PEG/silica
xerogels.

the release of diclofenac began immediately after immersion in
the medium. Such a trend was observed among all the irreg-
ular shaped-xerogels irrespective of the amount of drug added
( p < 0.05). At the end of 10 min of releasing, the cumulative
mean amount of diclofenac released was found to be 25 ± 2.1%.
However, the cumulative mean amount of diclofenac released
for pellets irrespective of the containing amount of diclofenac
was found to be 3 ± 1.9% for this time. This result suggests that
the burst release of diclofenac for the irregular-shaped xerogels
was affected by increasing the whole surface of xerogels exposed
to surrounding release medium compared with the pellets.
By examining the release profiles from all PEG xerogels dur-
ing 150 min of study, we see that the rate of the drug release
took place faster up to about 50 min, then the process got slower
and the rates significantly decreased, p < 0.05. Such a trend was
observed irrespective both of the amount of loaded diclofenac
in the matrices and the shape of matrices ( p < 0.05). Generally,
after 140 min of the experiment the maximum amount of di-
clofenac released—60 ± 4%, p < 0.05—was observed both for
the irregular-shaped xerogels and for the pellets irrespective of
the amount of loaded diclofenac. The uncompleted release of
drug molecules from the PEG/silica matrices suggests that dur-
ing the longer gel polymerization (in this experiments 5 days)
there is a covalent bond may be formed between the entrapped
reagent and the inner surface of the pores or interactions among
the drug and PEG and the xerogel matrices. The same conclu-
sions are described in the work of Brinker and Scherer (1990)
and Ro and Chung (1991).
The differences between the rates of drug released from
irregular-shaped and from pellets are not statistically significant
and this result suggests that the drug proceeds by the same mech-
anism of release. Most likely for all PEG/silica xerogels during
immersion time, a partial physical erosion of the material sur-
face by water and/or diffusion of water molecules into the pores
of the matrix took place. Following this process, the release of
the hydrophilic drug occurred in a controlled way through the
pores of the matrix. Additionally, to evaluate the mechanism of
drug released, three mathematic models of zero-order, first-order
and Higuchi are reported in Figure 6 and the correlation coeffi-
cients of obtained models for all diclofenac-loaded PEG/silica
xerogels are compared in Table 2. As can be seen from Figure
6, none of the PEG/silica xerogels followed a complete zero-
order release pattern. The correlation coefficient values for the
zero-order model in Table 2 also suggest the same result.
When the release data were analyzed according to the first-
order equation (log cumulative percentage of drug released ver-
sus time shown in Figure 6a), the correlation coefficients re-
ported in Table 2 exhibited a fair linearity. The best model for
the release study was expressed by Higuch’s equation (cumula-
tive percentage of drug release versus square root of time shown
in Figure 6b) and irrespective of the shape of xerogels, the cor-
relation coefficients r > 0.90 showed the highest linearity. The
obtained results suggest that diffusion is the rate-limiting mech-
anism of diclofenac release in this study.
TABLE 2
Correlation coefficients of different mathematical models for
diclofenac-loaded PEG/silica xerogels
Diclofenac-loaded
xerogel Zero-order (r ) First-order (r ) Higuchi (r )
Irregular-shaped 0.79 0.74 0.90
pellet 0.94 0.69 0.96
136
pellets
(180 ± 6 mg); r = 0.99
irregular particles
Degradation of xerogels [w/w %]
4 (355 ± 5 mg); r = 0.98
3,5
3
2,5
2 0
1,5
1
0,5
0
0 20 40
Time of degradation [days]
FIG. 7. Stability of PEG/silica xerogels during immersion time in SBF buffer solution. Dates were given as mean ± S.D. for n = 3. (a) Cumulative mean amount
of silica acid release from PEG/silica xerogels as a function of immersion time.
The rate of release of a drug having hydrophilic properties
also is influenced by the stability of matrices during immer-
sion time. Degradation of porous silica xerogel in an aqueous
medium occurs through hydrolysis of the siloxane bonds through
the entire network (Brinker and Scherer 1990). The amount of
dissolved silica acid during the immersion time in the water was
statistically constant for the same mass of xerogels and irrespec-
tive of the shape of xerogels used (results not shown).
Therefore in this work, the study of the effect of the whole
mass of matrices on the degradation of diclofenac-loaded
PEG/xerogels were evaluated. The variable masses 355 ± 5 mg
and 180 ± 6 mg of PEG/silica xerogels were compared. The cu-
mulative amount of silica acid dissolved and the degradation
rate of these xerogels over a period of 60 days are shown in
Figure 7 and 7a, respectively. After 60 days of the observa-
tion, 7 ± 1.7 mg of silica acid was dissolved and 2 ± 0.5%,
p < 0.05 of the gel mass was degraded for PEG/silica xerogel
with the average mass of 355 ± 5 mg. By examining the sta-
bility of matrix with the lowest mass, we found that 5 ± 1.2
mg of silica acid was dissolved and 3 ± 0.7% the gel mass un-
derwent degradation. The degradation rate of PEG/silica gels
seems to be independent of the total mass of the material used.
The obtained differences with the amounts of silica acid dis-
solved (∼2 mg) from different masses of xerogels are not dis-
tinctive. According to the literature on this subject, silica gels de-
grade by hydrolysis of the siloxane bonds on the surface through
the matrix, when water penetrates the porous structure result-
ing in mass loss of the silica gels (Kortesuo et al. 2001). The
more porous silica gels degrade much faster than smaller ones.
This was also shown in this study, because the microstructure
of silica xerogels is dependent on the TEOS/water ratio dur-
ing sol synthesis (Meixner and Dyer 1999). At the low value,
M. PROKOPOWICZ
Cumulative mass of silica acid
8 (a)
7
6
released [mg]
5
4
3
2
1
0 20 40 60 80
Time of degradation [days]
60 80
close to the hydrolytic stoichiometric value of 4, gelation occurs
readily and the structure of xerogels shows no mesoporosity or
microporosity. In this study, the PEG/silica xerogels were ob-
tained as the dense material with pores smaller than or equal to
1.5 nm.
Generally, the results of the rate of degradation of PEG/silica
xerogels revealed that the degradation for all PEG/silica gels,
irrespective of the mass and the shape of xerogel, followed a
zero-order kinetics (a linear relationship with a good correla-
tion coefficient r > 0.98 was observed). Furthermore, a com-
parison of stability of pure gel without the drug to stability
of gels with the drug revealed that stabilities in both cases
were not significantly different statistically ( p < 0.05, results
not shown). Based on these studies, the different drug con-
centrations did not have a statistically significantly influence
on the degradation rate of the PEG/silica gel matrix. In gen-
eral, a comparison of the results (Figures 6 and 7) reveals that
the drug release diffuses more rapidly from the matrix than
the resorption process taking place concurrently ( p < 0.05).
The release of diclofenac also was governed only by diffusion
process.
Additionally, Figure 8 illustrates a comparison of the FTIR
spectrum of PEG/silica xerogel before the drug release with
spectrum of the same matrix after the drug release. The spec-
trums are different only in the characteristic bands for PEG.
The bands at 2800–3000 cm−1 and at 1160–1060 cm−1 signifi-
cantly decreased in the spectra of matrix after the release study.
This result suggests that the hydrophilic molecules of PEG dis-
solved upon contact with water and diffused out of the xerogel
during the immersion time. This also suggests that the only in-
teractions involve weak hydrogen bonds between the PEG and
silanol groups in the silica lattice being formed.
 SILICA-POLYETHYLENE GLYCOL FOR DICLOFENAC RELEASE
100
80
60
%T
40
20
0
4000 3800 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000
Wavenumber[cm-1]
FIG. 8. FTIR spectra of PEG/silica xerogel before release of diclofenac and PEG/silica xerogel after release of diclofenac.
CONCLUSION
The monolithic with smooth surface PEG/silica xerogels
(crack-free) and crack, irregular-shaped xerogels loaded with
diclofenac were prepared using the sol-gel predoping process
with acetic acid as a catalyst. An indisputable advantage of
preparation of the drug matrix is the fact that the loading ef-
ficiency of diclofenac in all matrices is close to 100%, which
results from the simplicity of synthesis and lack of drug losses
at the gelation step. The rate of release of water-soluble di-
clofenac in vitro at 37◦ C, pH 7.4, from all monoliths is affected
only by the shape of xerogels used. The rates of drug released
from pellets and irregular-shaped xerogels were followed by
controlled and sustained release. However, the initial burst re-
lease was reported for the irregular-shaped xerogels. The study
of the stability of PEG/silica gels in the water solutions showed
that the rate of degradation followed a zero-order kinetics, irre-
spective of the shape, mass of xerogels, and the amount of drug
used.
In general, the release of the drug from all PEG-xerogels was
faster than the degradation rate of matrices, and this process
was controlled mainly by its diffusion rate through the pores
of the matrix. These results indicate that silica gels prepared by
synthesis with low-molecular PEG can be used as an implantable
drug delivery system in the different form (monolithic pellets or
particles) with a control release.
Future experiments can be designed to prepare xerogels
by using different pHs of the catalyst and different molar ra-
tio of precursors. Also, studies can use these PEG-xerogels
to encapsulate diclofenac and to study the rates of the drug
release.
137
PEG/silica xerogel
after drug-release
PEG/silica xerogel
before drug-release
800 600 400
REFERENCES
Ahola, M. S., Kortesuo, P., Kangasniemi, I., Kiesvaara, J., and Yli-Urpo, A. 2000.
Silica xerogel carrier material for controlled release of toremifene citrate. Int.
J. Pharmaceut. 195:219–227.
Ahola, M. S., Säilynoja, E. S., Raitavuo, M. H., Vaahtio, M. M., Salonen, J. I.,
and Yli-Urpo, A. 2001. In vitro release of heparin from silica xerogels. Bio-
materials 22:2163–2170.
Brinker, C. J. 1996. Porous inorganic materials. Curr. Opin. Solid State Mater.
Sci. 1:798–805.
Brinker, C. J., Keefer, K. D., Schaefer, D. W., Assink, R. A., Kay, B. D., and
Ashley, C. S. 1984. Sol-gel transition in simple silicates II. J. Non-Cryst.
Solids 63:45–59.
Brinker, C. J., and Scherer, G. W. 1990. The Physics and Chemistry of Sol-Gel
Processing. Boston: Academic Press.
Boinpally, R., Zhou, S., Poondru, S., Devraj, G., and Jasti, B. R. 2003. Lecithin
vesicles for topical delivery of diclofenac. Eur. J. Pharm. Biopharm. 56:389–
392.
Byrne, R. S., and Deasy, P. B. 2002. Use of commercial porous ceramic particles
for sustained drug delivery. Int. J. Pharm. 246:61–73.
Dunn, B., Miller, J. M., Dave, B. C., Valentine, J. S., and Zink, J. I. 1998.
Strategies for encapsulating biomolecules in sol-gel matrices. Acta Materialia
46(3):737–741.
Fang, J., Sung, K. C., Lin, H., and Fang, Ch. 1999. Transdermal iontophoretic
delivery of diclofenac sodium from various polymer formulations: in vitro
and in vivo studies. Int. J. Pharm. 178:83–92.
Gómez-Lechón, M. J., Ponsoda, X., O’Connor, E., Donato, T., Jover, R., and
Castell, J. V. 2003. Diclofenac induces apoptosis in hepatocytes. Toxicol. in
Vitro 17:675–680.
Guo, L., Hyeon-Lee, J., and Beaucage, G. 1999. Structural analysis of
poly(dimethylsiloxane) modified silica xerogels. J. Non-Cryst. Solids 243:61–
69.
Higginbotham, C. P., Browner, R. F., Jenkins, J. D., and Rice, J. K. 2003. De-
pendence of drying technique on surface area and pore size for polyethylene
glycol/tetramethoxysilicate hybrid gels. Mater. Lett. 57:3970–3975.
Koch, O. G., and Koch-Dedic, G. A. 1974. Siliconmolybdanblau-Verfahren. In
Handbuch der Spurenanalyse. p. 1105. Berlin: Springer.
138 M. PROKOPOWICZ
Kortesuo, P., Ahola, M., Kangas, M., Yli-Urpo, A., Kiesvaara, J., and Marvola,
M. 2001. In vitro release of dexmedetomidine from silica xerogel monoliths:
effect of sol-gel synthesis parameters. Int. J. Pharm. 221:107–114.
Kortesuo, P., Ahola, M., Karlsson, S., Kangasniemi, I., Yli-Urpo, A., and
Kiesvaara, J. 2000. Silica xerogel as an implantable carrier for controlled
drug delivery—evaluation of drug distribution and tissue effects after implan-
tation. Biomaterials 21:193–198.
Kramar, A., Turk, S., and Vrečer, F. 2003. Statistical optimisation of diclofenac
sustained release pellets coated with polymethacrylic films. Int. J. Pharm.
256:43–52.
Leonard, A., Blin, J. L., Jacobs, P. A., Grange, P., and Su, B. L. 2003. Chemistry
of silica at different concentrations of non-ionic surfactant solutoins: effect of
pH of the synthesis gel on the preparation of mesoporous silicas. Micropor.
Mesopor. Mat. 63:59–73.
Lin, J., and Brown, Ch. W. 1997. Sol-gel glass as a matrix for chemical and
biochemical sensing. Trends Anal. Chem. 16(4):200–211.
Liu, C. H., Kao, Y. H., Chen, S. C., Sokoloski, T. D., and Sheu, M. T. 1995.
In-vitro and in-vivo studies of diclofenak sodium controlled–release matrix
tablets. J. Pharm. Pharmacol. 47:360–364.
Livage, J. 1997. Sol-gel processes. Curr. Opin. Solid State Mat. Sci. 2:132–138.
Lowell, S., and Shields, J. E., 1991. Powder Surface and Porosity. London:
Chapman & Hall.
Meixner, D. L., and Dyer, P. N. 1999. Influence of sol-gel synthesis parameters
on the microstructure of particulate silica xerogels. J. Sol-Gel Sci. Technol.
14:223–232.
Morris, C. A., Rolison, D. R., Swider-Lyons, K. E., Osburn-Atkinson, E. J., and
Merzbacher, C. I. 2001. Modyfying nanoscale silica with itself: a method to
control surface properties of silica aerogels independently of bulk structure.
J. Non-Cryst. Solids 285:29–36.
Netz, D. J., Sepulveda, P., Pandolfelli, V. C., Spadaro, A. C., Alencastre, J. B.,
Bentley, M. V., and Marchetti, J. M. 2001. Potential use of gelcasting hy-
droxyapatite porous ceramic as an implantable drug delivery system. Int. J.
Pharm. 213:117–125.
Oberle, R. L., Das, H., Wong, S. L., Chan, K., and Sawchuk, R. J. 1984. Phar-
macokinetics and metabolism of diclofenac sodium in Yucatan miniture pigs.
Pharm. Res. 11:698–703.
Palumbo, G., Avigliano, L., Strukul, G., Pinna, F., Principe, D., Angelo, I. D.,
Annicchiarico-Petruzzeli, M., et al. 1997. Fibroblast growth and polymor-
phonuclear granulocyte activation in the presence of a new biologically active
sol-gel glass. J. Mater. Sci. Mater. Med. 8:417–421.
Prokopowicz, M., L
ukasiak, J., and Przyjazny, A. 2004. Utilization of a sol-
gel method for encapsulation of doxorubicin. J. Biomater. Sci. Polymer Edn.
15:343–356.
Prokopowicz, M., L
ukasiak, J., and Przyjazny, A. 2005. Synthesis and applica-
tion of doxorubicin–loaded silica gels as solid materials for spectral analysis.
Talanta 65:663–671.
Radin, S., Falaize, S., Lee, M. H., and Ducheyne, P. 2002. In vitro bioactivity
and degradation behavior of silica xerogels intended as controlled release
materials. Biomaterials 23:3113–3122.
Ro, J. C., and Chung, I. J. 1991. Structures and properties of silica gels prepared
by the sol-gel method. J. Non-Cryst. Solids 130:8–17.
Schwaiger, J., Ferling, H., Mallow, U., Wintermayr, H., and Negele, R. D. 2004.
Toxic effects of the non-steroidal anti-inflammatory drug diclofenac. Part I,
Aquatic Toxicol. 68:141–150.
Venkateswara Rao, A., Wagh, P. B., Haranath, D., Risbud, P. P., and Kumbhare,
S. D. 1999. Influence of temperature on the physical properties of TEOS silica
xerogels. Ceram. Int. 25:505–509.
Wilson, J., Pigott, G. H., Schoen, F. J., and Hench, L. L. 1981. Toxicol-
ogy and biocompatibility of bioglasses. J. Biomed. Mater. Res. 15:805–
817.
Witham, R. 1991. Voltaren (diclofenac sodium)-induced ileocolitis. Am. J. Gas-
troenterol. 86:246–249.