Densitometric Determination Diclofenac
Transcript of Densitometric Determination Diclofenac
Journal of Pharmaceutical and Biomedical Analysis28 (2002) 227–243
Densitometric determination of diclofenac,1-(2,6-dichlorophenyl)indolin-2-one and indolin-2-one in
pharmaceutical preparations and model solutions
Jan Krzek *, Malgorzała StarekDepartment of Inorganic and Analytical Chemistry, Collegium Medicum, Jagiellonian Uni�ersity, 9 Medyczna Str.,
30-688 Cracow, Poland
Received 25 July 2001; received in revised form 5 September 2001; accepted 25 September 2001
Abstract
A chromatographic-densitometric method for identification and quantitation of diclofenac and its impurities, i.e.1-(2,6-dichlorophenyl)indolin-2-one and indolin-2-one in pharmaceutical preparations and model solutions wasdeveloped. The effect of pH, temperature and ultra violet (UV) radiation on diclofenac’s concentration wasinvestigated. Chromatographic separation was performed on TLC silica gel coated plates with the mobile phase:cyclohexane–chloroform–methanol (12:6:1, v/v/v). Densitometric detection was carried out in UV at �=248 nm.The conditions for good separation and the detection limit were established. The recovery for diclofenac was 99.20%,for 1-(2,6-dichlorophenyl)indolin-2-one—92.34% and for indolin-2-one—95.85%. The method was used for qualityassessment of diclofenac in pharmaceutical preparations. Reliable results comparable to those determined by highperformance liquid chromatography (HPLC) were obtained. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Analysis of pharmaceuticals; Densitometry; TLC; Impurity; Diclofenac; 1-(2,6-dichlorophenyl)indolin-2-one; Indolin-2-one
www.elsevier.com/locate/jpba
1. Introduction
Diclofenac belongs to benzeneacetic acidderivatives and it is a sodium salt of 2-[(2,6-dichlorophenyl)-amino-phenyl]acetic acid [1]. It iscommonly used as a strong analgesic and anti-inflammatory agent in various drug formulationssuch as injections, suppositories and ointments[3].
Stability analysis has revealed that diclofenac[4,5] in acid solutions undergoes cyclization with aremoval of a water molecule according to thefollowing reaction: .
The reaction produces 1-(2,6-dichlorophenyl)-indolin-2-one, listed by the European Pharmaco-
* Corresponding author. Tel.: +48-12-657-0480; fax: +48-12-657-0262.
0731-7085/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S0731-7085(01)00622-7
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243228
poeia [1] and by the British Pharmacopoeia [2] asone of five potential impurities (imp. A) thatshould be taken into account in drug qualityassessment.
Available papers on determination of diclofe-nac and its potential impurities in commonmedicines were reviewed. Most authors recom-mend spectrophotometric methods for determina-tion of diclofenac and other constituentsoriginated from the matrix or for the analysis ofcomplex drugs [6–10].
Spectrofluorometric methods [11], X-rayfluorescence analysis [12,13], gas chromatography[14,15] and high performance liquid chromatogra-phy (HPLC) [16,17] were also used in the analysisof drugs containing diclofenac.
In this paper a new method for simultaneousdetermination of diclofenac, 1-(2,6-dichlorophenyl)indolin-2-one (imp. A) and in-dolin-2-one (imp. E) in various drug forms wasdeveloped. The method was used for theanalysis of drug stability in aqueous solutions,depending on pH, temperature and ultra violet(UV) radiation. This study is based on previoussuccessful applications [18–22] of densitometricmethod that can be an alternative to commonlyused HPLC.
2. Experimental
2.1. Apparatus
1. Densitometer TLC Scanner 3 with Cats4 soft-ware, manufacturer—CAMAG (Muthenz,Switzerland).
2. Sample applicator—Linomat IV, manufac-turer as above.
3. Computer—PC Pentium MMX, 16 MB RAM(Taiwan) and HP LaserJet 6L Printer (USA).
4. TLC Plates—9×15 cm (cut from 20×20 cmprecoated silica gel aluminium TLC sheetsArt. No. 16484, supplier, E. Merck—Darm-stadt, Germany).
5. TLC chamber of 18×9×18 cm in size, manu-facturer, Sigma-Aldrich.
6. Liquid Chromatograf LaChrom (Merck–Hi-
tachi) with detector DAD and automatic sam-ple feeder, computer controlled by theD-7000HSM software.
7. Filter, millipore corporation, Bedford, MA01730, pore size 0.45 �M.
2.2. Preparations examined
1. Majamil, tablets containing 50 mg of diclofe-nac; s. 1150101 exp. 012003 (Polpharma,Poland).
2. Diclofenac Sodium, tablets containing 50 mgof diclofenac; s. BNCAH43A exp. EXP0700(Norton, UK).
3. Rewodina retard, tablets containing 100 mg ofdiclofenac; s. 8004.1 exp. 11.2000 (AWD,Germany).
4. Voltaren, injections containing 75 mg of di-clofenac; s. MFD021999 exp. EXP022001(Novartis, Switzerland).
5. Olfen, injections containing 75 mg of diclofe-nac and 20 mg lidokaine; s. 9901404 exp.11.2003 (mepha, Switzerland).
6. Artotec, tablets containing 50 mg of diclofenacand 200 �g misoprostol; s. 856720 exp. 01.2001(Monsanto, France).
7. Dicloratio Retard, capsules containing 100 mgof diclofenac; s. 938117 exp. 08.2000 (Ratio-pharm, Germany).
8. Majamil, tablets containing 25 mg of diclofe-nac; s. 50898 exp. 082000 (Polpharma,Poland).
2.3. Reagents and chemicals
1. Standard substances: diclofenac (STADA); 1-(2,6-dichlorophenyl)indolin-2-one (Polpha-rma); indolin-2-one (Fluka).
2. Reagents: methanol (Merck); cyclohexane,chloroform (POCH Gliwice, Poland); phos-phoric acid (Park, UK); sodium dihydrogenphosphate (POCH Gliwice, Poland).
2.4. Solutions for densitometric analysis
1. Standard solutions: diclofenac 0.026%, w/v,1-(2,6-dichlorophenyl)indolin-2-one (imp. A)0.02%, w/v and indolin-2-one (imp. E) 0.01%,
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 229
w/v. Solutions were prepared in methanol–wa-ter (1:1, v/v).
2. Sample solutions:� tablets and capsules— ten tablets were pow-
dered in a mortar. Thoroughly powderedtablets containing approximately 20.0 mg ofdiclofenac were added to 20.0 ml ofmethanol–water (1:1, v/v) and shaken for30 min. The resulting suspension wasfiltered and used for further analysis.
� Injections—a volume of solution corre-sponding to 20.0 mg of diclofenac wasweighed and diluted with methanol–water(1:1, v/v) to 20.0 ml.
2.5. Solutions for HPLC analysis
1. Mobile phase: a mixture of 34 volumes of amixture of equal volumes of a 1 g/l solution ofphosphoric acid and a 1.6 g/l solution ofsodium dihydrogen phosphate adjusted to pH2.5, and of 66 volumes of methanol.
2. Impurity A standard solution: 0.002%, v/v inmobile phase.
3. Diclofenac standard solution: 0.005%, v/v inmobile phase.
4. Sample solutions for examination: an amountof preparation containing 50 mg of diclofenacwas weighed, shaken for 30 min in 70.0 ml of
Fig. 1. An example of densitogram and chromatogram of dileofenae (2), 1-2(2,6-dichlorophenyl)indolin-2-one (3) and indolin-2-one(1).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243230
Fig. 2. Absorption spectra for diclofenac (D), 1-(2,6-dichloropheny)indolin-2-one (A) and indolin-2-one (E) recorded fromchromatogram.
mobile phase, an than topped up to 100.0 mlwith the same solution. For tablets, the result-ing suspension was filtered through the mil-lipore filter of 0.45 �m.
3. Results
In further investigations the procedure was vali-dated to establish the conditions for determina-
Fig. 3. The relationship between peak area and concentration of, diclofenac (D; PD=10.258+94791×cD), 1-(2,6-dichloropheny)indolin-2-one (A; PA=245.96+81007×cA) and indolin-2-one (E; PE=122.97+148600×cE).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 231
tion of diclofenac and its potential impurities inpharmaceutical preparations [23].
The analysis was carried out by two analyticalchemists and the influence of previously plannedlittle changes of parameters on the results wastaken into account. The same apparatus andreagents were used.
3.1. Robustness (the effect of parameter changes onmeasurements)
To establish the optimal separation conditions,the amounts ranging from 1 to 10 �l of standardsolutions were applied in the form of 1 cm bandson the chromatographic plates of 9×15 cm.Chromatograms were developed in various mobilephases over different distances. A good separation
of constituents was reached over the mobile phasedistance of 12 cm in a relatively short time ofabout 60 min.
The UV densitometry was used for recordingthe chromatogram spots. The values of Rf forindividual constituents were determined fromdensitograms.
The use of mobile phase, cyclohexane–chloro-form–methanol (12:6:1, v/v/v) allows to obtainchromatograms with well developed and sepa-rated peaks (Fig. 1).
The absorption spectra within the range from200 to 400 nm recorded for each constituent arepresented in Fig. 2.
In the analysed absorption spectrum diclofenacshows a maximum at the wavelength �=280 nm,whereas the impurities at �=248 nm. The latter
Fig. 4. Chromatograms obtained by HPLC, standard impurity A (a) and analysed sample (b).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243232
Fig. 4. (Continued)
wavelength was chosen to determine all con-stituents because it allows to carry out researchsimultaneously and makes the analysis faster.
3.2. Linearity
For a linearity study seven solutions of differ-ent concentration ranging from 0.007 to 0.072%,v/v for diclofenac, from 0.005 to 0.050%, v/v forimpurity A and from 0.005 to 0.055%, v/v forimpurity E were prepared. The linear regressionmethod was employed (Fig. 3).
3.3. Quantitation limit and detection limit
The detection limit and the determination limitwere considered together, as under the established
chromatographic conditions well developed peakswith a negligible noise level were obtained. De-creasing concentrations of the analysed con-stituents were added to the preparation examinedwhile recording the peak areas on the chro-matograms. The limit of detection was determinedfrom the signal of S/N ratio equal to at least 4.The background was established from a ‘blind’sample. The following limits of detection werefound: 0.13 �g for diclofenac, 0.20 �g for impurityA and 0.05 �g for impurity E.
3.4. Selecti�ity
Since the aim of this study was to develop amethod for determining the active substance andits potential impurities in various drug forms, the
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 233
Table 1The results of diclofenac determination in preparations with statistical analysis
Preparation/declared concentration Determined concentration of Statistical analysis (n=7)of diclofenac diclofenac (%, w/w)
96.40Majamil/50 mg in tablets 95.80 ×=98.69102.20 S=4.69106.40 S×=1.77101.4093.60 �=98.69�4.33
95.00 R.S.D.=4.75
Diclofenac sodium/50 mg in tablets 103.4099.40 ×=100.8699.40 S=1.92103.20 S×=0.7398.80100.00 �=100.86�1.79
R.S.D.=1.90101.80
102.40Rewodina retard/100 mg in tablets 98.90 ×=99.1997.30100.60 S=2.29 S×=0.8795.7098.50 �=99.19�2.13
R.S.D.=2.31100.90
98.00Voltaren/75 mg in injections 102.13 ×=100.8099.73 S=2.57 S×=0.9799.87
105.47101.87 �=100.80�2.3798.53 R.S.D.=2.55
104.00Olfen/75 mg in injections ×=100.7298.67100.40 S=1.8299.87 S×=0.69
101.73 99.07 �=100.72�1.69101.33 R.S.D.=1.81
98.40Artotec/50 mg in tablets 95.80 ×=99.9198.60 S=3.25100.60 S×=1.2397.80103.00 �=99.91�3.01
105.20 R.S.D.=3.25
Dicloratio retard/100 mg in capsules 101.2096.30 ×=101.09101.00 S=4.12105.20 S×=1.56
96.50 �=101.09�3.82100.00R.S.D.=4.08107.40
99.20Majamil/25 mg in tablets 96.80 ×=100.1197.60 S=3.52103.60 S×=1.33
104.0096.00 �=100.11�3.25103.6 R.S.D.=3.52
×, arithmetic mean; S, standard deviation for individual points; S×, standard deviation for arithmetic mean; �, confidence intervalat 95% probability; R.S.D., relative standard deviation (%).
effect of potential impurities originated both fromthe matrix and from coexisting active substanceswas analysed. For this purpose model solutionscontaining comparative substances present in Ar-totec 50 and Olfen 75 and appropriate solutionsfor examination at concentrations identical tothose of model solutions were prepared.
The solutions were separated chromatographi-cally and the absorption spectra and peak areaswere recorded. Similar absorption spectra and
peak areas were obtained. It was found that addi-tional spots produced by coexisting substanceshave different Rf values (lidocaine Rf�0.13,misoprostol Rf�0.00, diclofenac Rf�0.32), thanthe impurities under investigation (imp. A Rf�0.67, imp. E Rf�0.21).
Thus, it was assumed that both the coexistingsubstances and potential impurities from the ma-trix had no effect on the determination of diclofe-nac and impurities A and E.
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243234
Tab
le2
The
resu
lts
ofdi
clof
enac
and
1-(2
,6-d
ichl
orop
heny
l)in
dolin
-2-o
ne(i
mpu
rity
A)
dete
rmin
atio
nfo
rva
riou
sco
ncen
trat
ions
ofH
Cl
atte
mpe
ratu
re22
°C
13da
ys20
days
43da
ysc
(HC
l)(m
ol/d
m3)
Pre
para
tion
0da
y5
days
i.AD
i.AD
i.AD
Di.A
Di.A
8.49
8.26
11.3
17.
027.
774.
9132
.90
4.48
1.00
0O
lfen
0.00
34.7
07.
698.
557.
707.
2822
.48
9.11
5.30
0.50
05.
8642
.75
0.00
22.2
70.
100
2.03
20.2
35.
1115
.41
8.89
11.6
90.
0041
.81
1.93
0.83
29.9
30.
9225
.87
2.65
20.7
05.
5412
.59
0.05
00.
0042
.25
0.00
26.3
20.
6626
.44
1.97
29.1
121
.66
69.9
20.
000.
020
0.00
62.1
60.
010
0.00
31.4
70.
0027
.17
0.26
24.2
60.
0082
.68
0.00
0.00
80.5
30.
0079
.18
0.00
71.3
30.
001
0.00
90.6
10.
0084
.37
8.72
2.92
11.2
42.
249.
845.
710.
001.
000
Dic
lora
tio
reta
rd5.
4521
.07
0.00
5.82
0.50
04.
965.
006.
663.
846.
311.
950.
0023
.48
3.16
0.94
11.2
81.
507.
513.
257.
253.
615.
750.
0026
.01
0.10
01.
3012
.86
2.00
10.8
92.
6023
.59
8.14
0.58
0.05
00.
0033
.86
0.79
25.3
90.
8421
.79
1.61
16.7
20.
020
0.00
43.8
00.
0043
.74
0.00
51.2
80.
0039
.69
1.12
65.6
823
.55
0.00
61.2
30.
000.
010
98.6
20.
001
0.00
101.
590.
0010
2.22
0.00
90.4
30.
0085
.41
0.00
11.4
52.
5015
.75
0.00
15.0
610
.35
0.00
1.00
0M
ajam
il6.
5111
.75
1.39
17.9
50.
500
10.8
36.
579.
306.
8819
.52
5.56
1.54
27.4
37.
511.
2125
.88
3.33
24.1
75.
2012
.06
8.99
6.04
0.10
00.
5730
.96
1.05
23.7
51.
6921
.76
5.27
26.7
815
.49
32.9
50.
600.
050
0.00
33.4
00.
020
0.00
35.1
70.
8430
.31
1.85
17.1
10.
0033
.19
0.00
0.00
0.01
029
.54
0.00
29.7
40.
0024
.09
0.00
28.9
70.
0029
.03
0.00
83.1
80.
0079
.60
0.00
85.1
467
.32
81.1
60.
001
0.00
0.00
i.A,
%(w
/w)
conc
entr
atio
nof
impu
rity
Aw
ith
refe
renc
eto
acti
vesu
bsta
nce
(dic
lofe
nac)
;D
,di
clof
enac
conc
entr
atio
n(%
,w
/w).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 235
Tab
le3
The
resu
lts
ofdi
clof
enac
and
1-(2
,6-d
ichl
orop
heny
l)in
dolin
-2-o
ne(i
mpu
rity
A)
dete
rmin
atio
nfo
rva
riou
sco
ncen
trat
ions
ofH
Cl
atte
mpe
ratu
re37
°C
3da
ys4
days
7da
ys9
days
14da
ysc
(HC
l)(m
ol/d
m3)
Pre
para
tion
0da
y1
day
i.AD
i.AD
i.AD
i.AD
Di.A
Di.A
Di.A
10.7
324
.78
13.0
110
.34
33.3
711
.23
43.1
55.
9945
.72
5.30
49.1
80.
000.
00O
lfen
34.7
01.
000
13.0
216
.95
22.7
017
.34
32.3
318
.54
37.6
933
.28
12.9
00.
500
50.1
77.
596.
7742
.75
0.00
44.4
20.
100
6.36
35.2
78.
0626
.35
12.2
727
.64
13.9
017
.69
20.1
916
.76
0.00
41.8
12.
340.
9548
.72
3.72
33.3
72.
7831
.71
6.35
31.7
18.
1730
.31
11.5
926
.09
0.00
42.2
50.
050
2.88
41.9
51.
5343
.02
2.53
42.8
03.
1948
.33
35.4
769
.92
6.32
34.3
50.
000.
020
0.00
72.4
10.
010
0.00
57.8
40.
0058
.15
0.27
57.0
60.
4753
.73
0.73
51.1
80.
0082
.68
0.00
0.00
79.6
60.
0077
.79
0.00
78.4
90.
0080
.69
0.00
10.
0076
.29
0.00
90.6
10.
0090
.73
9.96
3.93
24.2
02.
6316
.29
1.50
14.8
56.
041.
011.
000
14.1
60.
00D
iclo
rati
ore
tard
4.73
21.0
70.
007.
690.
500
6.92
5.09
7.92
4.74
8.80
3.15
8.55
2.33
10.7
90.
000.
0023
.48
2.89
0.10
015
.03
1.78
10.5
32.
238.
412.
738.
913.
847.
385.
266.
670.
0026
.01
0.78
1.66
14.9
31.
9313
.65
2.36
10.5
22.
8825
.44
9.58
0.05
04.
6710
.22
0.00
33.8
60.
0038
.89
0.02
01.
0731
.37
1.35
30.8
51.
4624
.56
1.85
25.6
03.
1821
.81
0.00
43.8
00.
000.
0058
.50
0.00
49.9
30.
3932
.65
0.42
34.6
30.
010
0.67
31.8
30.
0061
.23
0.00
60.1
30.
0094
.14
0.00
107.
500.
0088
.89
0.00
96.0
999
.20
0.00
10.
0010
6.09
0.00
85.4
10.
007.
111.
000
10.7
05.
2510
.14
5.57
14.4
61.
5611
.56
0.00
20.2
20.
001.
3911
.75
4.13
Maj
amil
5.98
6.24
6.69
6.13
9.45
4.10
8.82
3.17
13.5
51.
750.
500
1.54
27.4
32.
087.
421.
8410
.66
2.49
10.9
63.
689.
633.
5217
.70
7.08
30.9
66.
546.
700.
710.
100
0.57
24.8
50.
050
1.33
16.5
21.
4315
.63
2.44
11.8
62.
1811
.27
5.69
12.1
50.
0032
.95
0.43
0.00
24.4
40.
0024
.36
1.01
21.5
70.
8320
.12
2.40
18.1
50.
020
0.00
33.1
90.
0026
.93
0.00
37.3
90.
0029
.44
0.00
34.7
70.
0049
.34
27.6
30.
000.
0031
.15
0.01
00.
0028
.97
0.00
0.00
178
.20
0.00
69.7
20.
0074
.97
0.00
81.7
30.
0079
.32
0.00
81.1
60.
0076
.96
i.A,
%(w
/w)
conc
entr
atio
nof
impu
rity
Aw
ith
refe
renc
eto
acti
vesu
bsta
nce
(dic
lofe
nac)
;D
,di
clof
enac
conc
entr
atio
n(%
,w
/w).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243236
Tab
le4
The
resu
lts
ofdi
clof
enac
and
1-(2
,6-d
ichl
orop
heny
l)in
dolin
-2-o
ne(i
mpu
rity
A)
dete
rmin
atio
nfo
rva
riou
sco
ncen
trat
ions
ofH
Cl
atte
mpe
ratu
re60
°C
3da
ysc
(HC
l)(m
ol/d
m3)
4da
ys0
day
1da
y2
days
Pre
para
tion
i.AD
i.AD
i.AD
i.AD
i.AD
51.5
10.
0052
.74
0.00
35.6
034
.41
0.00
1.13
34.7
00.
001.
000
Olf
en21
.57
4.04
47.7
10.
0053
.11
0.00
26.9
10.
000.
500
0.00
42.7
542
.06
20.1
250
.60
10.4
331
.72
26.2
74.
3716
.69
0.10
00.
0041
.81
30.4
123
.94
41.9
823
.54
28.8
718
.00
0.05
00.
0042
.25
14.5
929
.28
17.1
727
.45
22.8
226
.62
18.5
140
.49
25.9
07.
3169
.92
0.00
0.02
01.
7261
.04
3.91
43.1
15.
8139
.70
6.14
39.4
70.
010
0.00
82.6
80.
5689
.58
0.89
84.8
90.
4589
.79
73.4
10.
2290
.61
0.00
0.00
11.
9396
.39
2.47
71.6
73.
3766
.96
0.00
010.
0010
1.74
0.57
100.
551.
7976
.69
2.45
69.8
93.
0097
.83
64.1
80.
7910
3.96
0.00
0.00
001
1.54
77.2
82.
3570
.18
3.03
65.2
90.
0000
001
0.00
104.
460.
8110
1.28
13.2
30.
0013
.53
0.00
13.5
60.
000.
0010
.24
Dic
lora
tio
reta
rd1.
000
0.00
21.0
710
.40
2.91
11.6
11.
2311
.54
0.00
0.50
00.
0023
.48
8.26
3.52
8.04
4.41
11.4
83.
2711
.60
7.95
0.00
2.89
26.0
10.
000.
100
7.85
9.29
10.8
64.
830.
050
10.4
10.
000.
0033
.86
1.98
9.22
6.67
22.8
98.
3819
.22
12.1
731
.33
7.68
43.8
01.
550.
020
0.00
37.2
10.
010
1.13
33.8
82.
8633
.76
3.97
22.2
90.
0061
.23
0.67
104.
830.
001
0.67
100.
031.
7110
7.66
2.94
104.
030.
0085
.41
0.00
1.92
97.1
32.
0797
.14
2.16
105.
7796
.68
0.00
103.
310.
000.
0001
110.
280.
0000
11.
5410
7.78
1.45
105.
982.
1097
.71
0.00
107.
410.
001.
1294
.48
1.94
94.8
92.
0510
0.30
93.9
20.
000.
0000
001
0.00
110.
9730
.88
0.00
19.7
60.
0021
.59
Maj
amil
0.00
1.00
01.
3911
.75
17.5
20.
0027
.46
0.00
17.5
70.
0020
.91
3.67
0.00
19.7
827
.43
1.54
0.50
021
.91
0.10
06.
6111
.89
7.03
23.1
31.
550.
5730
.96
28.7
49.
2612
.73
12.4
514
.88
6.46
23.9
215
.15
6.51
0.05
015
.91
32.9
50.
0029
.59
0.02
02.
7821
.02
5.13
17.0
512
.32
11.8
80.
0033
.19
9.71
50.4
90.
010
2.35
36.7
31.
0832
.76
3.65
33.4
60.
0028
.97
3.24
0.35
91.9
61.
0787
.98
3.35
101.
2689
.20
0.00
81.1
60.
000.
001
0.00
101.
452.
8293
.17
2.44
93.1
35.
0891
.44
0.00
010.
0098
.30
1.24
96.8
11.
5593
.35
2.67
100.
3490
.85
0.00
0.00
001
0.00
99.4
40.
980.
0000
001
91.7
71.
5089
.61
1.29
86.8
40.
0010
2.47
0.00
94.2
6
i.A,
%(w
/w)
conc
entr
atio
nof
impu
rity
Aw
ith
refe
renc
eto
acti
vesu
bsta
nce
(dic
lofe
nac)
;D
,di
clof
enac
uco
ncen
trat
ion
(%,
w/w
).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 237
3.5. Accuracy
The accuracy of the method was defined as%recovery of analyte added from 80 to 120% ofrelevant substances with respect to preparationsexamined. The following values of recovery werefound for individual constituents (for n=3), di-clofenac: × , 99.20%; relative standard deviation(R.S.D.), 0.61% �95%, 99.20%�1.51; impurity A:× , 92.34%; R.S.D., 1.35% �95%, 92.34%�3.10;impurity E:× , 95.85%; R.S.D., 4.18% �95%,95.85%�9.98.
The results obtained were used for developingthe method of quantitation for furtherinvestigations.
3.6. Precision
The precision of the method was expressed as aconsistence degree between the results of analysescarried out repeatedly. The tests were conductedfor a model mixture solution containing diclofe-nac (0.025%), impurity A (0.023%) and impurityE (0.055%). To determine the results consistencymeasurements of peak areas for each constituentwere done. The scatter of results was character-ised by the standard deviation (S.D.) which alsoconfirms reproducibility of the method and theR.S.D. in compliance with the arithmetic mean(× ) from six measurements. For each constituentthe following results were obtained, diclofenac:from 3492.0 to 3780.1, × , 3630.6; S, 108.5;R.S.D., 2.99%; impurity A: from 2718.8 to 2810.3,× , 2756.7; S, 32.6; R.S.D., 1.18%; impurity E:from 1481.6 to 1587.2, × , 1523.3; S, 39.9;R.S.D., 2.62%.
3.7. Quantitati�e analysis
Standard solutions, 4 �l of diclofenac and im-purity A, 2 �l of impurity E and preparationsolutions of 1 �l for determination of diclofenacand of 20 �l for impurities were applied in theform of 1 cm bands on chromatographic plates of9×12 cm. Chromatograms were developed overthe mobile phase distance of 12 cm by using themobile phase, cyclohexane–chloroform–methanol (12:6:1, v/v/v), and then dried at room
temperature. The densitograms were recorded atthe wavelength of �=248 nm.
The concentrations of each constituent inpreparations examined were determined by com-paring the peak areas for appropriate standardand sample solutions.
The results of diclofenac determination are pre-sented in Table 1. The results for impurities havenot been included, as no impurities were found.
To confirm the lack of impurities, HPLC wasused as a comparative method [2].
3.8. Determination of impurity A by HPLC
Standard solutions and corresponding prepara-tion solutions of 20 �l each were applied sequen-tially on the column of 250×4 mm, packed withoctadecylosilane chemically bound on the surfaceof amorphous silica (Lichrosorb RP-18) and theflow rate of mobile phase was fixed to 1 ml permin. The detection was made in UV at �=254nm.
Individual peaks on chromatograms were iden-tified by comparing the retention times for stan-dard solutions and for the solutions examined.The peak areas were used for determining impu-rity A.
Examples of chromatograms for standard solu-tion of impurity A and preparation solution areshown in Fig. 4.
Similar chromatograms were obtained for allpreparations and no peaks at the retention timefor impurity A (ca. 7.7 min) were noted. The onlysingle strong peak at the retention time for di-clofenac (ca. 12.3 min) was visible on allchromatograms.
The results have confirmed the reliability ofdata obtained by the chromatographic-densito-metric method which also revealed no impurities.
3.9. The effect of pH and temperature on concen-trations of diclofenac and impurities in model solu-tions of selected drugs
The examination was carried out on the follow-ing drugs: Dicloratio retard 100 (capsules), Olfen75 (injections) and Majamil 50 (tablets). The solu-tions for examination were prepared from pow-
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243238
dered tablets by weighing 20 mg of diclofenac ortaking an appropriate volume of injections. Afterdissolving in 20 ml of hydrochloric acid at con-centrations ranging from 1 to 10−7 mol/l (pHfrom 0 to 7), the samples were diluted withmethanol (in 1:1, v/v) at time intervals specified inTables 2–4. The chromatographic-densitometricmethod was employed. The chromatogram spotswere identified based on the absorption spectraand Rf values.
The areas of appropriate peaks were used forquantitation. The results are presented in Tables2–4.
A decreasing concentration of impurity A wasfound after incubation of diclofenac solutions at60 °C on day 4. This can indicate, for example,decomposition of that impurity. Additional exam-inations were carried out and this hypothesis wasconfirmed by the appearance of an additionalpeak at Rf�0.95.
3.10. The effect of UV on concentrations of diclofe-nac and impurities in model solutions
The drugs mentioned above were examined.Sample solutions were prepared from powderedtablets by weighing an amount corresponding to20 mg of diclofenac or taking an appropriatevolume of injection. After dissolving in 20 ml ofmethanol–water (1:1, v/v) and filtering, the sam-ples were exposed to UV at �=254 and 366 nmat time intervals specified in Table 5. Thechromatogram spots were identified by densito-metric measurements. The results are presented inFig. 6.
The chromatograms of samples indicated thepresence of diclofenac and two additional spotsoriginated from substances of Rf values differentfrom those of the impurities (A and E) examined.Thus, it was concluded that after UV irradiation,new compounds denoted as Y and Z, appeared.These compounds have not been analysed in de-tail due to the lack of appropriate standard sub-stances. The quantitation was confined only todetermination of concentrations for diclofenacand other compounds by using the internal nor-malisation, from the following formula, % i=xi/�x, where, % i, constituent concentration i [%]; xi,
peak area for constituent under examination; �x,sum of peak areas. The results are shown in Table5.
4. Discussion
A new method that combines TLC and densito-metry was developed for simultaneous identifica-tion and quantitation of the active substance andits impurities in order to assess the quality ofdiclofenac and impurities 1-(2,6-dichlorophenyl)-indolin-2-one (impurity A) and indolin-2-one (im-purity E) and fulfilling the requirements specifiedin the European and British Pharmacopoeias. Inthis study two of five impurities listed in thepharmacopoeias were analysed [2,3].
The conditions for separation of diclofenac andanalysed impurities and the method for chromato-graphic identification were established by usingappropriate standard solutions.
It was found experimentally that the mobilephase, cyclohexane–chloroform–methanol (12:6:1, v/v/v) provided good constituent separation,and at the same time gave well developed peakson densitograms what is of utmost importance indensitometric analysis. The constituents differ notonly in their position on the chromatograms (di-clofenac Rf�0.32, impurity A Rf�0.67, impurityE Rf�0.21) but also in the absorption spectra interms of specific absorbance maximum and shape.These parameters appear to be of a great signifi-cance in quantitation and identification (Fig. 2).
The method presented above meets all generalrequirements of good laboratory practice in termsof specificity, with respect to potential matrixconstituents, originated from vehiculum or otheractive substances in complex medicines. Underestablished conditions selective determination ofdiclofenac, as well as impurities appearing to-gether and in the presence of other constituentssuch as misoprostol (Rf�0.00) and lidokaine(Rf�0.13) was possible. No influence of othersubstances originated from the matrix in the stud-ied pharmaceutical preparations obtained fromdifferent manufactures was observed. The methodis characterised by good precision, very high sen-sitivity and constituent detection (diclofenac 0.13
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 239
Tab
le5
The
resu
lts
ofde
term
inat
ion
for
dicl
ofen
acan
dpr
oduc
tsof
its
deco
mpo
siti
onun
der
UV
irra
diat
ion
(con
cent
rati
ons
calc
ulat
edby
the
inte
rnal
norm
alis
atio
nm
etho
dde
scri
bed
inth
epa
per)
15m
in30
min
45m
in60
min
75m
in90
min
105
min
120
min
Det
erm
inat
ion
prod
uct
Pre
para
tion
0�
(nm
)
80.4
580
.44
76.1
9O
lfen
74.7
925
470
.08
68.3
0D
iclo
fena
c96
.62
93.3
182
.87
7.96
8.41
10.5
512
.17
14.1
36.
9115
.28
2.99
3.38
Y10
.23
11.8
311
.14
13.2
613
.03
15.7
916
.42
Z0.
003.
6991
.47
88.8
286
.32
84.6
382
.41
81.7
779
.06
75.6
0D
iclo
fena
c36
696
.62
8.39
8.42
8.78
10.9
912
.16
6.85
13.8
13.
385.
95Y
4.33
5.29
6.95
8.81
7.25
8.78
10.5
9Z
0.00
2.57
Dic
lora
tio
reta
rd87
.19
82.6
482
.26
79.1
475
.65
90.1
673
.07
95.2
810
0D
iclo
fena
c25
43.
906.
889.
578.
7411
.02
12.4
0Y
0.00
0.71
2.57
8.91
10.4
89.
5712
.12
13.3
27.
2714
.53
3.86
0.00
Z93
.74
366
91.1
689
.51
87.2
487
.92
85.0
683
.78
Dic
lofe
nac
100
98.0
92.
753.
954.
755.
716.
457.
88Y
0.00
0.00
1.91
6.08
6.53
8.01
6.38
8.49
4.35
8.34
1.91
0.00
Z94
.12
254
92.7
792
.36
90.4
587
.73
86.5
270
.40
Dic
lofe
nac
100
95.0
9M
ajam
il2.
542.
804.
165.
525.
8513
.94
Y0.
002.
042.
304.
694.
845.
406.
757.
633.
5715
.67
0.00
2.87
Z90
.22
366
88.4
085
.99
82.9
180
.20
78.2
074
.63
Dic
lofe
nac
100
93.5
03.
534.
357.
259.
0510
.02
12.5
0Y
0.00
1.43
2.84
8.06
9.66
9.84
10.7
511
.78
12.8
76.
94Z
5.07
0.00
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243240
�g, impurity A 0.20 �g, impurity E 0.05 �g), widerange of linearity (0.007–0.072%, w/v for diclofe-nac; 0.005–0.050%, w/v for impurity A; 0.005–0.055%, w/v for impurity E) and high recovery(99.20%, w/v for diclofenac; 92.34%, w/v for im-purity A; 95.85%, w/v for impurity E).
It is important to point out that UV measure-ments at the analytical wavelength �=248 nmcommon to all constituents gave correct and re-peatable results.
The examinations carried out on selected prepa-rations were confined only to the determination ofdiclofenac, as no impurities were found. The re-sults presented in Table 1 are characterised byhigh accuracy, repeatability and precision, asconfirmed by statistical analysis. For diclofenacdetermination, the confidence intervals for eachpreparation ranging from 1.69 to 4.33% andR.S.D. from 1.81 to 4.75% were obtained. Somedoubts as to the suitability of the method couldarise mainly from the negative results obtained for
impurity A, the level of which has been specifiedin the European Pharmacopoeia. The results ob-tained for the model mixture based on peak areasmeasurements in quantitative analysis confirmgood precision of the method. The calculatedvalues of S.D. and R.S.D. are comparable, whatsuggests that the results obtained under the estab-lished conditions of determination for each con-stituent are of a comparable quality. Thus, anadditional determination of impurity A was car-ried out by HPLC, as recommended in the phar-macopoeia. The reliability of the densitometricmethod has been confirmed, as the obtained chro-matograms showed only peaks at the retentiontime typical of diclofenac (ca. 12.3 min). For allpreparations examined no peaks at the retentiontime typical of impurity A (ca. 7.7 min) werefound.
In further investigations, based on the pub-lished reports [5], the effect of pH in aqueoussolutions on diclofenac decomposition was
Fig. 5. An example densitogram for impurity A (A) and unidentified product (N) arisen after 4-day incubation at 60 °C. The peakmarked with U is the front of a mobile phase.
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 241
Fig. 6. An example of densitogram and chromatogram of diclofenac (D) and two unidentified compounds (Y and Z) obtained afterUV sample irradiation (a), and absorption spectra of both products recorded directly from chromatogram (b).
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243242
analysed. When determining diclofenac and impu-rity A (as no traces of impurity E were detected)it has been found that the decomposition of di-clofenac is faster in solutions with lower pH anddepends on the temperature.
The fastest decomposition of diclofenac wasobserved in solutions of high acidity. The higherpH, the slower decomposition of diclofenac. Anincreasing temperature considerably speeds up thedecomposition process. A temperature rise up to60 °C led to a complete decomposition of diclofe-nac within 24 h (Tables 2–4).
The changes in diclofenac concentration wereaccompanied by a proportional increase in theconcentration of impurity A, depending on pHand temperature of the solution. After 4-day incu-bation at 60 °C (Table 4) some changes resultingprobably from decomposition of impurity A wereobserved. The concentration of impurity A de-creased in comparison to that of three previousdays (Fig. 5).
In addition, the effect of UV on the stability ofdiclofenac was analysed (Fig. 6). Densitogramsshowed both the peaks generated by diclofenac(Rf�0.32), as well as unidentified substances de-noted as Y (Rf�0.19) and Z (Rf�0.23). As theproduct of diclofenac decomposition remains un-known, the internal normalisation was used forquantitation to determine both diclofenac and theproducts of its decomposition.
The examinations carried out at �=254 and366 nm have indicated no significant differences inthe decomposition process. In both cases similarphenomena were observed. The results listed inTable 5 show decreasing concentration of diclofe-nac after UV irradiation, thus indicating gradualdecomposition induced by UV. The changes indiclofenac concentration were accompanied bycorresponding changes in concentrations of Y andZ. These products of diclofenac decompositiondiffer from impurities A and E both in peakpositions and in the absorption spectra (Fig. 5).
5. Conclusions
A chromatographic-densitometric method forsimultaneous determination of diclofenac and 1-
(2,6-dichlorophenyl)indolin-2-one and indolin-2-one was developed. Based on the results presentedabove it has been demonstrated that the methodcan be used as an alternative to HPLC, both indiclofenac purity assessment, as well as in kineticanalysis.
The developed method is characterised by highsensitivity, selectivity and accuracy. This methodis worth recommending because of a short time ofanalysis, little usage of reagents and simplicity.
It has been also shown, that diclofenac decom-poses in aqueous solutions releasing 1-(2,6-dichlorophenyl)indolin-2-one (impurity A), theconcentration of which depends on pH and tem-perature of the solution.
An adverse effect of UV on the stability ofdiclofenac in model solutions was found. UVirradiation reduced the concentration of diclofe-nac, while generating new products of differentphysical and chemical properties.
This finding appears to be interesting for tech-nology not only in terms of drug formulation butalso in therapy, especially in the case of drugsadministered orally as they are exposed to theacid environment of the stomach.
References
[1] European Pharmacopoeia, Third Edition, Strasbourg1997.
[2] British Pharmacopoeia 2000, Comlete EditionCD.[3] J.K. Podlewski, A. Chwalibogowska-Podlewska, Leki
Wspołczesnej Terapii, Split Trading Sp. Z o.o, Warsaw,1994 In polish.
[4] A. Avgerinos, T. Karidas, S. Malamataris, J. Chro-matogr. Biomed. Appl. 130 (1993) 324–329.
[5] M. Zajaac, B. Stanisz, W. Musiał, Acta Polon. Pharm. 55(1998) 371–374.
[6] T. Perez-Ruiz, C. Martinez-Lozano, A. Sanz, M.T. San-Miguel, J. Pharm. Biomed. Anal. 16 (1997) 249–254.
[7] S. Agatonovic-Kustrin, L. Zivanovic, M. Zecevic, D.Radulovic, J. Pharm. Biomed. Anal. 16 (1997) 147–153.
[8] M. El Sadek, Egypt. J. Pharm. Sci. 32 (1991) 457–463.[9] A.K. Shakya, G.K. Joshi, P. Mishra, Indian J. Pharm.
Sci. 54 (1992) 44–46.[10] P. Parimoo, A. Bharathi, R.J. Babu, C.N. Gupta, S.
Sudha-Rao, East. Pharm. 40 (1997) 111–112.[11] L.A. Carreira, M. Rizk, Y. El-Shabrawy, N.A. Zakhari,
S.S. Toubar, J. Pharm. Biomed. Anal. 13 (1995) 1331–1337.
J. Krzek, M. Starek / J. Pharm. Biomed. Anal. 28 (2002) 227–243 243
[12] M.A. Raggi, F. Lucchini, P. Da Re, Boll. Chim. Farm.132 (1993) 29–31.
[13] M.A. Raggi, P. Da Re, F. Lucchini, F. Parisi, A. Salvi, J.Pharm. Biomed. Anal. 8 (1990) 975–978.
[14] R.T. Sane, S.R. Surve, M.G. Gangrade, V.V. Bapat, N.L.Chonkar, Indian Drugs 30 (1993) 66–72.
[15] C.S. Sastry, A.R. Rao, C.V. Gopala-Krishna, A.G.Murthy, Indian J. Pharm. Sci. 50 (1988) 175–178.
[16] R.T. Sane, R.S. Samant, V.G. Nayak, Drug. Dev. Ind.Pharm. 13 (1987) 1307–1314.
[17] J.L. Chawla, R.A. Sodhi, R.T. Sane, Indian Drugs 33(1996) 171–178.
[18] J. Krzek, Acta Polon. Pharm. 55 (1998) 99–103.[19] J. Krzek, M. Starek, J. Planar Chromatogr. 12 (1999)
356–360.[20] J. Krzek, A. Kwiecien, J. Pharm. Biomed. Anal. 21 (1999)
451–457.[21] J. Krzek, A. Maslanka, Acta Polon. Pharm. 57 (2000)
23–26.[22] J. Krzek, M. Starek, J. AOAC Int., in press.[23] The Validation of Analytical Procedure Used in the Ex-
amination of Pharmaceutical Materials WHO/PHARM.89.541/Rev.2 Genf, 1989.