Stability of diclofenac sodium in the inclusion complex with β-cyclodextrin in the solid state

Article abstract of DOI:10.1211/0022357991776930

The aim of this study was to characterize the thermal stability of diclofenac sodium both alone and in the inclusion complex with β-cyclodextrin in the solid state, by determination of the number of the products of its decomposition, which were identified by GC-MS. The molar ratio of diclofenac sodium in the inclusion complex with β-cyclodextrin was 1:1. The decomposition of diclofenac sodium both alone and in inclusion complex with β-cyclodextrin occurred according to the first-order reaction. The HPLC of the samples thermostated at 80°C gave five products of decomposition, which were identified by GC-MS. Diclofenac sodium in the inclusion complex with β-cyclodextrin was more thermally stable. Thermal decomposition of diclofenac sodium leads to formation of five products, of which 4-chloro-10H-9-acridinone had not been reported previously in the literature.

Full text of DOI:10.1211/0022357991776930

J. Pharm. Pharmacol. 1999, 51: 1213±1218
Received April 29, 1999
Accepted June 7, 1999
# 1999 J. Pharm. Pharmacol.
Stability of Diclofenac Sodium in the Inclusion Complex
with b-Cyclodextrin in the Solid State
BARBARA CWIERTNIA, TERESA HLADON AND MACIEJ STOBIECKI*
Department of Inorganic and Analytical Chemistry, Karol Marcinkowski Un� � iꢀersity of Medical Sciences,
Grunwaldzka 6, 60780 Poznan and *Institute of Birganic Chemistry, Polish Academy of Sciences,
Noskowskiego 12/14, 61704 Poznan, Poland
Abstract
The aim of this study was to characterize the thermal stability of diclofenac sodium both
alone and in the inclusion complex with b-cyclodextrin in the solid state, by determination
of the number of the products of its decomposition, which were identi®ed by GC±MS.
The molar ratio of diclofenac sodium in the inclusion complex with b-cyclodextrin was
1
: 1. The decomposition of diclofenac sodium both alone and in inclusion complex with
b-cyclodextrin occurred according to the ®rst-order reaction. The HPLC of the samples
thermostated at 80 C gave ®ve products of decomposition, which were identi®ed
by GC±MS.
ꢀ
Diclofenac sodium in the inclusion complex with b-cyclodextrin was more thermally
stable. Thermal decomposition of diclofenac sodium leads to formation of ®ve products, of
which 4-chloro-10H-9-acridinone had not been reported previously in the literature.
Non-steroidal anti-in¯ammatory drugs (NSAIDs),
such as diclofenac sodium, are often used in the
treatment of rheumatic diseases (Brogden et al
and in inclusion complex with b-cyclodextrin in the
solid state, determining the number and identity of
its decomposition products, as well as explaining
the mechanism of thermal decomposition of this
substance.
1980; Orienti et al 1991; Kaiser & Pineda 1997).
The mechanism of action of NSAIDs involves
inhibition of prostaglandin cyclooxygenase neces-
sary for metabolism of arachidonic acid (Van
Bruchhausen et al 1993). NSAIDs can cause a
number of undesirable effects such as ulcerogenic
or in some cases hepatotoxic effects. Stability of
these drugs is of utmost importance as the products
of their decomposition may enhance the un-
desirable effects.
Recent literature has brought many reports on the
physical and chemical properties of diclofenac
sodium (Clarke 1986; Sato et al 1997), and its
pharmacological effects (Sagara et al 1992). How-
ever, little information has been given on its stability.
One of the methods used to increase the stability of a
pharmaceutical is through the formation of its
inclusion complex with a cyclodextrin (Szejtli 1988;
Duch e ne & Wouessidjewe 1990; Connors 1995).
The aim of this study was to characterize the
thermal stability of diclofenac sodium both alone
Materials and Methods
Materials
b-Cyclodextrin was from Chinoin (Budapest,
Hungary). Diclofenac sodium, acetonitrile and
methanol (HPLC purity) were purchased from
Sigma Chemical Co. (St Louis, MO). Deuterated
dimethylsuphoxide (DMSO-d ) (99Á95%) was
6
obtained from Merck KGaA (Darmstadt, Ger-
many). All other reagents were of analytical grade.
Preparation of the diclofenac sodium±b-cyclodex-
trin inclusion complex
The inclusion complex of diclofenac sodium with
b-cyclodextrin was prepared by the kneading
method. b-Cyclodextrin, 1Á135 g, was wetted with
ethanol in an agate mortar and kneaded to form a
paste. Then 0Á318 g diclofenac sodium and ethanol
were added. The sample was kneaded for
approximately 60 min and dried to constant mass at
Correspondence: B. Cwiertnia, Department of Inorganic and
Analytical Chemistry, Karol Marcinkowski University of
Medical Sciences, Grunwaldzka 6, 60780 Poznan, Poland.

1
214
BARBARA CWIERTNIA ET AL
ꢀ
solid state by X-ray, DSC and by C NMR.
1
05 C. The ®nal product was characterized in the
Thermal stability of pure diclofenac sodium
and that present in the inclusion complex with
b-cyclodextrin
13
Thermal stability of diclofenac sodium and its
inclusion complex with b-cyclodextrin was studied
at 60, 70 and 80 C in dry air. The vials containing
samples of either 0Á0250 g diclofenac sodium or
0Á1142 g inclusion complex were placed in heated
chambers placed in sand baths at the given tem-
peratures.
At certain time intervals the contents of the vials
were quantitatively transferred to 25-mL measuring
¯asks, which were made-up with water to the mark.
After ®ltering, the concentration of diclofenac
sodium was determined spectrophotometrically.
Instrumentation
ꢀ
Characterization of the diclofenac sodium±b-
cyclodextrin complex in the solid state was per-
formed by X-ray using an X-ray diffractometer
TUR M62. X-ray analysis was performed using
Ê
nickel-®ltered CuK radiation (l ˆ 1Á5418 A). The
a
voltage and current were 30 kV and 30 mA,
respectively. The inclusion complex obtained was
identi®ed by comparing its X-ray patterns with
those of pure b-cyclodextrin, diclofenac sodium
and a physical mixture of b-cyclodextrin and
diclofenac sodium.
DSC analysis was performed for diclofenac
sodium, b-cyclodextrin, the physical mixture of the
two at the molar ratio of 1 : 1, and the product of
kneading. Thermograms were obtained by use of a
Shimadzu DSC-50 instrument using vented alumi-
nium pans. The analysis was performed in a
nitrogen atmosphere. All samples were run at a
HPLC of the decomposition products of diclofenac
sodium alone and in the inclusion complex with
b-cyclodextrin
Analysis of the decomposition products of pure
diclofenac sodium and diclofenac sodium in the
inclusion complex with b-cyclodextrin, stored at
ꢀ
� 1
ꢀ
ꢀ
scanning rate of 10 C min , from 20 to 480 C
Figure 2A±D).
80 C, was performed by an HPLC method (Kubala
(
et al 1993; Mason & Hobbs 1995). A high-per-
formance liquid chromatograph (Shimadzu Co.,
Kyoto, Japan) was used with a stationary phase of
LiChrosorb RP-18. The mobile phase was a mix-
ture of acetonitrile±water (50 : 50, v/v), pH adjus-
ted to 3Á3 with acetic acid. The detector wavelength
was set at 276 nm, the ¯ow rate of the mobile phase
13
C NMR spectra for analysis were taken on a
Varian Uniti 300 spectrometer using the following
13
parameters: C frequency 75Á43 MHz, frequency
range 6000 MHz, memory 65 kB, pulse width
1
Spectroscopic measurements of the samples of
1Á3 ms, acquisition time 0Á999 s, resolution 0Á3 Hz.
�
1
� 1
0
Á3 mol L
concentration were performed in
DMSO-d . The chemical shifts were determined
was 1Á5 mL min and the column temperature was
ꢀ
6
20 C.
with respect to the signal of the solvent (DMSO-d6
39Á500 ppm) and expressed for TMS, which was the
internal standard.
GC±MS analysis of the decomposition products of
diclofenac sodium alone and in inclusion complex
with b-cyclodextrin
Determination of diclofenac sodium in the
inclusion complex with b-cyclodextrin
Before the GC±MS studies, the samples were
separated into fractions by extraction to solid state.
Two millilitres aqueous solution of the inclusion
Samples of the complex (20 mg) were placed in
measuring ¯asks, and then distilled water was
added up to the volume 25 mL. The contents were
mechanically shaken for 30 min. The concentration
of diclofenac sodium in the solution was deter-
mined spectrophotometrically (Agatonovic-Kustrin
et al 1997; P e rez-Ruiz et al 1997) at lmax ˆ 275 nm
by use of a Cary 118 Varian UV-vis spectro-
photometer. The amount of diclofenac sodium in
the inclusion complex with b-cyclodextrin was
calculated as the difference between the mass of
weighed portions of the complex studied and the
mass of diclofenac sodium determined spectro-
photometrically in the solution. The amount of
diclofenac sodium contained in 1-g inclusion
complex with b-cyclodextrin was 0Á2189 g.
�
1
complex (11Á4 g L ) or diclofenac sodium
�
1
(2Á5 g L ) was placed onto the column packed
with octadecylsilane phase C-18, previously con-
ditioned with 5 mL methanol. The elution was
performed with: 2 mL 1 : 1 methanol±water solu-
tion; 2 mL 4 : 1 methanol±water solution; 2 mL
methanol.
The presence of the decomposition products in
the eluate was searched for by a TLC method on
plates coated with F254 silica gel (0Á25-mm thick).
The fractions collected were dried at room tem-
perature in a nitrogen atmosphere to dry mass. The
dry mass was dissolved in acetone and then 20 mL
of this prepared solution was injected onto the
column.

DICLOFENAC SODIUM±b-CYCLODEXTRIN INCLUSION COMPLEX
1215
The parameters of the gas chromatograph
Hewlett-Packard 5890 series H, USA) were as
follows. The capillary column was a DB-5
Results and Discussion
(
For the X-ray diffraction results the Debye re¯ec-
tions were measured in the y angle range from 2 to
0 . The spectrum of b-cyclodextrin (Figure 1B)
when compared with diclofenac sodium (Figure
A) showed a higher number of re¯ections of
higher intensity. The interplanar distances corre-
(
30 m 6 0Á25 mm) Folson Ca (USA), the carrier gas
ꢀ
2
�
1
was helium at a ¯ow rate of 1 mL min , and the
detector was a Hewlett Packard 5971 A mass
spectrometer (USA).
1
sponding to the high intensity re¯ections were
Ê
1
9Á56, 14Á35, 9Á96, 7Á60 and 6Á02 A. The X-ray
diffraction pattern of diclofenac sodium revealed
fewer high intensity re¯ections and they corre-
sponded to the following interplanar distances:
Ê
1
3Á31, 10Á43, 9Á81 and 9Á33 A. The pattern of the
physical mixture of diclofenac and b-cyclodextrin
Figure 1C) showed the peaks characteristic of both
(
Ê
b-cyclodextrin (19Á56, 9Á96, 7Á60 A) and diclofenac
Ê
sodium (13Á31, 9Á33 A). The X-ray diffraction pat-
tern of the kneading product (Figure 1D) was poor
in re¯ections in the angle range of 15±20 , which
ꢀ
testi®ed to a reduced ordering of the crystal lattice.
The fragment of the pattern corresponding to the
other angle range revealed the peaks characteristic
of the inclusion complex which corresponded to the
Ê
interplanar distances of 12Á05, 8Á86, 4Á80 A. Their
appearance indicated the formation of a new crystal
Figure 2. Differential scanning calorimetry curves of diclo-
fenac sodium±b-cyclodextrin system. A, Diclofenac sodium;
B, b-cyclodextrin; C, physical mixture of diclofenac sodium
and b-cyclodextrin; D, diclofenac sodium±b-cyclodextrin
inclusion complex.
Figure 1. X-ray diffraction patterns. A, diclofenac sodium; B,
b-cyclodextrin; C, physical mixture of diclofenac sodium and
b-cyclodextrin; D, diclofenac sodium±b-cyclodextrin inclu-
sion complex.

1
216
BARBARA CWIERTNIA ET AL
lattice of the inclusion complex of diclofenac
sodium and b-cyclodextrin.
The position of the carbonyl carbon was shifted
from 175Á733 (diclofenac sodium) to 175Á686 ppm
(inclusion complex), so Dd ˆ 0Á047 ppm. More-
over, the position of the carbon atom of the
The formation of such an inclusion complex was
also con®rmed by DSC studies. The thermogram of
diclofenac sodium (Figure 2A) showed a single
CH � COONa group was shifted from 44Á339 ppm
2
ꢀ
endothermic peak at 288Á78 C, corresponding to
(diclofenac sodium) to 43Á871 ppm (inclusion
complex), so Dd was 0Á468 ppm. The changes
suggested that the cavities in b-cyclodextrin were
®lled with the acetate groups of diclofenac sodium
and that the inclusion complex was formed.
the melting point of this compound. Between 283±
2
ꢀ
85 C diclofenac sodium begins to melt and
ꢀ
decompose, so the peak at 285 C pointed to the
beginning of the decomposition process. The ther-
mogram of b-cyclodextrin (Figure 2B) showed two
To determine the thermal stability, the samples
ꢀ
ꢀ
ꢀ
endothermic peaks at 95 C and 321Á84 C, the for-
mer corresponding to the water evaporation from b-
cyclodextrin cavities, the latter corresponding to
the decomposition of the b-cyclodextrin molecule.
In the thermogram of the inclusion complex of
diclofenac sodium and b-cyclodextrin (Figure 2D),
the endothermic peaks characteristic of b-cyclo-
dextrin and diclofenac sodium disappeared,
were stored at 60, 70 and 80 C for 24 months and
the effect of these temperatures was analysed. The
results are presented as semilogarithmic plots of
ln c ˆ f (t), where c is parent compound con-
centration and t is time (Figure 4). The results
proved that the decomposition of diclofenac
sodium both alone and in inclusion complex with
b-cyclodextrin occurred according to the simple
equation of the ®rst-order reaction.
The kinetic parameters of the decomposition
reaction, given in Table 1, proved that the thermal
stability of diclofenac sodium stored in each of the
three temperatures was much higher when it was in
the form of inclusion complex with b-cyclodextrin.
This conclusion was supported by the thermo-
dynamic parameters given in Table 2.
ꢀ
whereas the exothermic peak at 259Á28 C, which
did not occur in any other thermograms, appeared.
The presence of this peak was evidence for the
formation of the inclusion complex.
Another method used to con®rm the identity of
the product obtained by the kneading of b-cyclo-
13
dextrin and diclofenac sodium was C NMR
spectroscopy (Figure 3). The spectrum of the
kneading product revealed changes in the chemical
shift d of the carbon atoms relative to their posi-
tions in the spectrum of pure diclofenac sodium.
To determine the number of decomposition pro-
ducts, the two forms of diclofenac sodium stored at
ꢀ
an elevated temperature (80 C) were subjected to
1
3
Figure 3.
complex.
C NMR spectra of diclofenac sodium (A), b-cyclodextrin (B), and diclofenac sodium±b-cyclodextrin (C) inclusion

DICLOFENAC SODIUM±b-CYCLODEXTRIN INCLUSION COMPLEX
1217
Table 1. Kinetic parameters of diclofenac sodium alone and
in inclusion complex with b-cyclodextrin in solid state.
Kinetic parameters
9
Temperature K 10 k (s
� 1
� 1
� 1
)
t0 (d ) t (d )
Á1
0Á5
Diclofenac
sodium
333
343
53
3Á264Æ 0Á07 441Á0
4Á701Æ 0Á14 313Á7
6Á147Æ 0Á21 213Á4
2899Á6
2062Á5
1402Á8
3
Inclusion
complex
333
343
53
3Á072Æ 0Á06 511Á6
3Á821Æ 0Á05 364Á7
4Á896Æ 0Á14 236Á8
3364Á1
2397Á9
1557Á3
3
HPLC analysis. The chromatograms of the samples
revealed the presence of ®ve decomposition pro-
ducts (t ˆ 1Á996, 5Á900, 8Á510, 14Á050, 26Á307).
R
The retention time of diclofenac sodium was
1
2Á315 min and that of b-cyclodextrin was
13Á367 min (Figure 5).
The same number of decomposition products was
obtained for uncomplexed diclofenac and for that
in the inclusion complex with b-cyclodextrin. This
suggested that the mechanism of its decomposition
was the same for the conditions studied. This meant
that b-cyclodextrin did not interfere with the reac-
tion of diclofenac sodium decomposition.
ꢀ
Figure 5. Chromatograms of samples stored at 80 C for 24
Figure 4. Thermal degradation of diclofenac sodium as a
function of time at 60 (A), 70 (B) and 80 C (C). d Diclofenac
sodium, s inclusion complex.
months obtained by HPLC. A. Diclofenac sodium uncom-
plexed and B. diclofenac sodium±b-cyclodextrin inclusion
complex.
ꢀ

1
218
BARBARA CWIERTNIA ET AL
Table 2. Thermodynamic parameters for the degradation reaction of diclofenac sodium alone and in the inclusion
complex with b-cyclodextrin in the solid state.
Thermodynamic parameters
�
1
)
� 1
DS (J mol� 1ÁK� 1
a
lnA
r
E
a
(J mol
DH (J mol
)
)
Diclofenac sodium � 3728Æ 1472 � 8Á33Æ 4Á29 � 0Á9980 30996Æ 12235 28228Æ 12235
Inclusion complex � 2738Æ 848 � 11Á38Æ 2Á48 � 0Á9988 22765Æ 7057 19997Æ 7057
� 315Æ 36
� 340Æ 21
a, slope; lnA, coef®cient frequency; r, regression coef®cient; E , activation energy; DH, enthalpy; DS, entropy.
a
n ˆ 6; a ˆ 0Á05.
Clarke, E. G. C. (1986) Isolation and Identi®cation of Drugs in
Pharmaceuticals, Body Fluids, and Post Mortem Material.
Second edn, The Pharmaceutical Press, London
Connors, K. A. (1995) Population characteristics of cyclo-
dextrin complex stabilities in aqueous solution. J. Pharm.
Sci. 84: 843±848
To identify the products of decomposition, they
were subjected to GC±MS analysis. The substances
were separated into fractions by extraction to solid
phase. The greatest number of compounds more
hydrophobic than diclofenac sodium was found in
the fraction extracted with 50% of methanol.
The chromatograms obtained as a result of gas
chromatographic separation showed ®ve peaks
corresponding to the products of diclofenac sodium
Duch e ne, D., Wouessidjewe, D. (1990) Pharmaceutical uses of
cyclodextrins and derivatives. Drug Dev. Ind. Pharm. 16:
2
487±2499
Kaiser, P. K., Pineda, R. M. D., Corneal Abrasion Patching
Study Group. (1997) A study of topical nonsteroidal anti-
in¯ammatory drops and no pressure patching in the treat-
ment of corneal abrasions. Ophthalmology 104: 1353±1359
Kubala, T., Gambhir, B., Borst, S. I. (1993) A speci®c stability
indicating HPLC method to determine diclofenac sodium in
raw materials and pharmaceutical solid dosage forms. Drug
Dev. Ind. Pharm. 19: 749±757
Mason, J. L., Hobbs, G. J. (1995) A rapid high performance
liquid chromatographic assay for the measurement of diclo-
fenac in human plasma. J. Liq. Chromatogr. 18: 2045±2058
Orienti, I., Fini, A., Bertasi, V., Zecchi, V. (1991) Inclusion
complexes between non-steroidal antiin¯ammatory drugs
and b-cyclodextrin. Eur. J. Pharm. Biopharm. 37: 110±112
P e rez-Ruiz, T., Martinez-Lozano, C., Sanz, A., San Miguel,
M. T. (1997) Flow extraction spectrophotometric method for
the determination of diclofenac sodium in pharmaceutical
preparations. J. Pharm. Biomed. Anal. 16: 249±254
decomposition: t ˆ 15Á58, 16Á86, 17Á21, 19Á67 and
R
2
0Á34 min. The mass spectra of the products were
recorded and the products were identi®ed on the
basis of the mass of the molecular ion and the
fragment ions. Four of the ®ve identi®ed com-
pounds were described by Sun & Fabre (1994).
Those authors presented a general scheme of
diclofenac sodium decomposition. The compounds
we found were 1-(2,6-dichlorophenyl)oxindole,
1
-(2,6-dichlorophenyl)isatin, 2-[(2,6-dichloro-phenyl)
amino]benzyl alcohol, and N-(2,6-dichlorophenyl)
anthranililaldehyde, the same as those found by
Sun & Fabre (1994). Our ®fth compound was 4-
chloro-10H-
-acridinone, to the best of our knowledge not
reported previously in the literature as a product of
diclofenac sodium decomposition.
Sagara, K., Nagamatsu, Y., Yamada, I., Kawata, M., Mizuta,
H., Ogawa, K. (1992) Bioavailability study of commercial
sustained-release preparations of diclofenac sodium in gastro-
intestinal physiology-regulated dogs. Chem. Pharm. Bull.
9
4
0: 3303±3306
Sato, H., Miyagawa, Y., Okabe, T., Miyajima, M., Sunada, H.
1997) Dissolution mechanism of diclofenac sodium from
(
wax matrix granules. J. Pharm. Sci. 86: 929±934
Sun, S. W., Fabre, H. (1994) Practical approach for validating
the TLC assay of an active ingredient in a pharmaceutical
formulation. J. Liq. Chromatog. 17: 433±445
Szejtli, J. (1988) Cyclodextrin Technology. J. E. D. Davies
Kluwer Acad. Publ., Dordrecht
Van Bruchhausen, F., Ebel, S., Frahm, A. W., Hackenthal, E.
(1993) Hagers Handbuch der Pharmazeutischen Praxis.
Springer-Verlag Berlin, Heidelberg
References
Agatonovic-Kustrin, S., Zivanovic, L., Zecevic, M., Radulo-
vic, D. (1997) Spectrophotometric study of diclofenac-
Fe(III) complex. J. Pharm. Biomed. Anal. 16: 147±153
Brogden, R. N., Heel, R. C., Pakes, G. E., Speight, T. M.,
Avery, G. S. (1980) Diclofenac sodium: a review of its
pharmacological properties and therapeutic use in rheumatic
diseases and pain of varying origin. Drugs 20: 24±48