Fresenius' Journal of

Fresenius J Anal Chem (1992) 342:723 -728

......

@ Springer-Verlag1992

Determination of sulfamethoxazole and trimethoprim by ratio spectra derivative spectrophotometry J. J. Berzas Nevado, J. M. Lemus Gallego, and G.Castafieda Pefialvo Department of Analytical Chemistry and Foods Technology, University of Castilla-La Mancha, E-13071 Ciudad Real, Spain Received June 15, 1991; revised September 10, 1991

Summary. A new method is described for analysing the binary mixture of sulfamethoxazole and trimethoprim, using the ratio spectra derivative spectrophotometry. The procedure does not require any separation step. Calibration graphs were linear up to 25 mg/1 of sulfamethoxazole and up to 50 mg/1 of trimethoprim; their 1:5 and 5:1 (m/m) mixtures can be resolved with recoveries between 97 and 104%. The method was satisfactorily applied for determining sulfamethoxazole and trimethoprim in different pharmaceutical products.

this method, the absorption spectrum of the binary mixture is divided (amplitude to amplitude) by a standard spectrum of one of the chromophores and the first derivative of the ratio spectrum is obtained. A calibration graph is obtained before. The method has been already tested for different mixtures, salicylaldehyde and its 3-hydroxybenzaldehyde and 4-hydroxybenzaldehyde isomers [11, 12] and, recently, for sulfamethizole and nitrofurantoine [13]. The mixture of SMX and TMP has been resolved in samples containing up to 25 mg/1 of SMX or up to 25 mg/1 of TMP in the 1:5 and 5:1 (m/m) ratios, also in three different pharmaceutical products, Septrim, Eduprim and Mucorama, with recovery values between 97 and 102%.

Introduction Sulfonamides are widely used in medicine and veterinary practice as antibacterial drugs. Many sulfonamides are now available as pharmaceutical products and numerous methods have been developed for their determination. Different chromatographic techniques have been reported for the analysis of sulfonamide mixtures [ 1 - 5]. Methods for the determination of the total content of sulfonamides have been also reported; these methods involving spectrophotometric techniques based on the Bratton-Marshall procedure [6] are widely studied and have been automated by using an airsegmented continuous flow analyser [7] and by flow injection analysis [8]. Recently, derivative spectrophotometry (on the basis of the colour obtained by Bratton-Marshall reaction) has been used for determining the total content of sulfonamides in urine and honey without pretreatment of the samples [9]. Nowadays, the pharmaceutical products containing sulfonamides consist only of one drug or one sulfonamide mixed with another drug that operates for increasing the power of the sulfonamide, i.e., sulfaquinoxaline and pyrimethamine, sulfadiazine and trimethoprim, sulfamethoxazole and trimethoprim, etc. In these cases of mixtures containing few chromophores with overlapped spectra, derivative spectrophotometry can be a useful technique for resolving the samples without previous separation steps and without previous chromogenic reaction. In this work, the mixture of sulfamethoxazole (SMX) and trimethoprim (TMP) is investigated and resolved by using the first derivative of the ratio spectra method [10]. In Offprint requests to." J. J. Berzas Nevado

Experimental Apparatus

A Beckman Instrument DU-70 Spectrophotometer connected to an IBM-PS computer fitted with Beckman Data Leader Software [14] and an Epson FX-850 printer, was used for all the measurements and treatment of data. Solutions

Analytical grade solvents were used. Trimethoprim and sulfamethoxazole were obtained from Sigma Chemical Company Products, and their stock solutions were prepared in a concentration of 250 mg/1 in 50% (v/v) ethanol/water. Ammonium chloride/ammonia buffer solution 0.5 tool/1 (pH 10) was made from analytical grade-reagent. Procedure Synthetic mixtures. They were prepared in 25 ml calibrated flasks containing between 4 - 2 0 mg/1 of SMX and/or between 4 - 2 0 rag/1 of TMP, and 5 ml of buffer solution pH 10, and diluted with water and ethanol to the mark (the resulting final solution is 10% in ethanol). The absorption spectra were recorded against a reagent blank and stored in the IBM-PS computer. For determining TMP, the stored spectra of the mixtures were divided by a standard spectrum of SMX of 15 rag/1. The ratio spectra thus obtained were smoothed by the use of 15 experimental points, and the first derivatives were calculated with A 2 = 5 rim. The amplitudes at 293 nm

724 (1DD293), 244 nm (aDD244) and their peak-to-peak measurement of signals from the maximum at 244 nm to the minium at 293 nm, (tDD293, 244) were proportional to the concentration of TMP. For determining SMX, the stored spectra of the mixtures were divided by a standard spectrum of TMP of 15 rag/1. The ratio spectra thus obtained were smoothed by the use of 15 experimental points, and the first derivatives calculated with A 2 = 5 nm. The amplitudes at 2 6 4 n m (aDD264), 252.5 nm (1D252.5) , and their peak-to-peak (1DD264, 252.5) measurements were proportional to the concentration of SMX.

Commercial pharmaceutical preparations. The method was applied for resolving the mixtures in three pharmaceutical preparations: Eduprim, Mucorama and Septrim. For each medicament studied, one tablet was weighed and finely powdered. An accurately weighed portion of the powder, equivalent to about 80 mg of SMX and 16 mg of TMP, was extracted with 80 ml of ethanol by occasional shaking for 1 h. The mixture was filtered (filter paper) into a 100-ml calibrated flask, the residue washed several times with the same solvent and diluted to volume. A portion of 25 ml was diluted to 100 ml with 50% (v/v) ethanol/water. After an aliquot of 4 ml had been carried into 25-ml calibrated flask, 5 ml of buffer solution was added and diluted with water and ethanol to the mark (the resulting final solution was 10% in ethanol). The absorption spectra were recorded against a reagent blank and stored in a IBMPS computer. For determining SMX and TMP, the absorption spectra were handled as in the general procedure for synthetic mixtures. Results and discussion

0.600

250

nm

Ld 0 . 5 0 0

ZO 0 . 4 0 0

0.300

::: ::: :~

<: 0.200

*"*'*'~ * *

0,100

0.000 0

I

I

I

I

/

I

2

4

6

8

10

12

4

pH Fig. 1. Influence of pH on the 10 mg/1 ([~) SMX and (*) TMP absorption spectra 1.00

0.80

I,I E) Z 0.60 <2 133 rY C) (13 03 0 . 4 0

0.20

Development of the method The method proposed was successfully used for the analysis of sulfonamides in pure form and in commercial pharmaceutical preparations. SMX and TMP were dissolved in 50% (v/v) ethanol/water and their solutions were stable for 15 days at least. The changes on absorption spectra with the pH were studied between pH 2 and 13 (Fig. 1). The TMP spectra were unchanged between p H 2 and 6 and between pH 9 and 13. The SMX spectra were unchanged between pH 2.5 and 4.5 and between pH 13 and 7.5. A basic medium was considered as suitable and an ammonium chloride/ammonia buffer solution p H 10 was used. The ethanol content of the medium slightly affects the absorbances of both compounds, and as a result 10% was selected as optimum. Under these conditions, SMX and TMP in diluted solutions (-~ 10-3mol/1) were stable for 2 h at least. The absorption spectra of SMX and TMP and their mixture in the 3 2 0 - 200 nm wavelength region are shown in Fig. 2. As can be seen, the spectra of the compounds overlap sufficiently and a mathematical treatment of the data is recommended for resolving the mixture. In this way, the proposed methodology gave very good results. To optimize the simultaneous determination of the SMX and TMP by using the ratio spectra derivative spectrophotometry, it is necessary to test the influence of the variables: divisor standard concentration, A2 and smoothing function. All these variables were studied.

0.00 200

~

j

,

2,50

260

290

" 320

WAVELENGT,H (nm) Fig. 2. Absorption spectra of SMX (4 mg/1), TMP (4 mg/1) and their mixture

For determining SMX, the effect of TMP concentration divisor on the intensity of SMX ratio spectra was studied and a concentration of 15 mg/1 was considered as suitable. The influence of the A2 for obtaining the first derivative was tested and A2 = 5 was selected as optimum value. Due to the extent of the noise levels on the ratio spectra, a smoothing function was used on the basis of Savitzky and Golay method [15] and 15 experimental points were considered as suitable. The ratio spectra and their first derivative ratio spectra obtained under these conditions are shown in Fig. 3. Calibration graphs were made by measuring the amplitude from the maximum at 264 nm, from the minimum at 252.5 nm, and from peak-to-peak. For determining TMP an analogous procedure was followed. Figure 4 shows the ratio spectra of different standards of TMP and their first derivative. The concentration of the divisor (SMX) was 15 mg/1, the first derivative was made with A2 = 5 nm and the derivative ratio spectra

725

Fig. 3. Ratio spectra (a) and first derivative (b) for different concentrations o f S M X : I (2 m g 1), 2 (5 nag/l), 3 (10 mg/1), 4 (15 mg/1), 5 (25 rag/l) and 6 (35 mg/1)

Fig. 4. Ratio spectra (a) and first derivative (b) for different concentrations o f T M P : I (2 m g 1), 2 (10 rag/l), 3 (20 mg/1), 4 (30 mg/1), 5 (40 rag/l) and 6 (50 rag/l)

Table 1. Statistical data for calibration graphs in the determination o f T M P and S M X Equations

Regression coefficient

Range (mg/1)

S t a n d a r d deviation Slope 10-4

Intercept 10-3

TMP 1DD244 = 1.2.10 -3 + 5.94.10 -3 C a 1DD293 = 1.11.10 3 + 13.26.10-3 C a 1DD244, 293 2.38" 10 -3 + 19.19-10 .3 C a 1D255.75 = 1.2.10 . 4 + 8.2.10 . 4 C" t D 2 2 5 . 5 = 2 . 4 " 1 0 4 + 1.66.10-3 C" 1D27o.25 = 0.3' 10 - 4 -}- 0.8" 10 4 C a =

0.9999 0.9999 0.9999 0.9999 0.9993 0.9997

1 --30 1 --50 I --30 1 --50 1--15 1 --35

0.29 0.29 0.64 0.02 0.31 0.01

0.46 0.62 1.03 0.04 0.25 0.01

0.9999 0.9999 0.9999 0.9999 0.9999 0.9999

1 --25 1 --25 1 --25 1 --25 1 --23 1 --23

1.75 1.23 2.97 0.04 0.02 0.01

2.31 1.63 3.93 0.05 0.04 0.01

SMX IDD252.5=-2.2"10 4+0.05109-ca 1DDz64 = 2.9- 10 4 + 0.36484.C a 1DD252.5,264 = 0.68" 10 . 4 q- 0.08758 C a 1D259. 5 = 0.4-10 - 4 q- 9.2.10 . 4 C" 1D28v.5 = -- 1.0- 10 - 4 q- 8.6" 10-4 C a 2D3o1.75 = 0 . 3 - 1 0 - 4 + 1.7-10 4 C a a C o n c e n t r a t i o n in mg/1

726 were s m o o t h e d with 15 experimental points. Calibration graphs were m a d e from the m a x i m u m at 293 nm, from the m i n i m u m at 244 nm, and from peak-to-peak.

Statistical study In Tables 1 and 2 are summarized the most characteristic statistical d a t a obtained from the different calibration graphs and from the reproducibility of the reagent b l a n k and a s t a n d a r d o f 15 mg/1 S M X or T M P by ten successive scans. In all cases g o o d results were obtained. F o r the p r o posed method, the best limit quantification for S M X was obtained from p e a k - t o - p e a k (t D252.5, 264) and the best limit quantification for T M P was obtained from the m i n i m u m at 244 rim.

Determination of SMX and TMP in synthetic mixtures Some binary mixtures of S M X and T M P were m a d e from stock solutions in relations from I : 5 to 5:1 and were resolved by the p r o p o s e d method. Table 3 shows the results o f the

determinations of different mixtures. The recoveries were between 100% and 101% for S M X and between 97% and 104% for T M P for all wavelengths studied. In the case o f the determination of T M P , the best recoveries were obtained from peak-to-peak, here the recoveries were between 99% and 102%. These results show that the m e t h o d is effective for the simultaneous determination o f S M X and T M P by the first derivative o f the ratio spectra. The results obtained for the determination o f S M X and T M P mixtures in commercial pharmaceutical preparations are shown in Table 5. The percentages o f recovery obtained were between 99% and 101% for S M X in all products at all wavelengths studied. In the case of T M P , recoveries between 101% and 100% for Septrim and 101% and 102% for E d u p r i m were found. W h e n T M P was determined in M u c o r a m a , satisfactory results were only obtained at 244 nm. The high values obtained from the m i n i m u m at 293 nm (and consequently from the peak-to-peak) can be imputed to the presence o f iodopropylideneglycole in the pharmaceutical c o m p o u n d . In the rest o f the cases (Septrim and Eduprim) the matrices did not affect the a b s o r p t i o n spectra o f the mixture studied.

Table 2. Statistical parameters for the determination of 10 mg/1 of TMP and SMX. Replicate n = 10 0.015

Signal measured

Standard deviation (mg/1)

Relative error (%)

Identif. limit (mg/1)

TMP 1DD244 tDD293 1DD244, 293 tD225.5 1D255.75 2D27o.25

0.239 0.131 0.125 0.467 0.141 0.548

4- 1.20 _ 0.67 _+0.64 + 0.02 4- 0.58 ± 0.02

0.03 0.21 0.15 0.23 >0.03 >0.03

TMP

Quantif. limit (mg/1)

0.11 0.70 0.49 0.77 >0.03 >0.03

0,010

u~ > .~ >_ ~c~ cz I

*

SMX *

•

MIXTURE

0.005

0.000

0")

C~ la_

SMX 1DD25z.5 1DD264 aDD252.5,264 1D259. 5 1D287. 5 ZD301.vs

-0.005 0.085 0.087 0.088 0.104 0.073 0.280

4- 0.42 4-_0.42 Jr- 0.43 ± 0.52 4- 0.36 4- 1.43

0.06 0.14 0.05 >0.03 0.17 >0.03

0.21 0.47 0.17 >0.03 0.56 >0.03

225.5 nm -0.010

259.5 nm 255.75 nm T

220

287.5 nm

T

250 280 WAVELEN©T,H, ( n m )

q-

,310

Fig. $. First derivative for 4 mg/1 of SMX, TMP and their mixture

Table 3. Results obtained in the determination of TMP and SMX in synthetic mixtures by using the proposed method Composition of mixture

TMP recovery (%)"

TMP (mg/1)

SMX (rag/l)

1D244

1DD293

1DD244. 293

1DD2s2.s

1DD264

tDD252.5. 264

20 16 12 8 4 4 4 4 4

4 4 4 4 4 8 12 16 20

100 100 100 100 97 98 97 97 98

102 101 101 102 101 100 103 101 104

101 101 100 102 99 99 101 100 102

101 101 101 100 100 101 100 100 100

100 100 100 100 100 101 100 100 100

100 100 100 100 100 101 100 100 100

" Average of two determinations

SMXrecovery (%)a

727 Table 4. Results obtained in the determination of TMP and SMX in synthetic mixtures by using the first and second derivative spectra

TMP (rag/l) 20 16 12 8 4 4 4 4 4

SMX (mg/1)

TMP recovery (%)"

4 4 4 4 4 8 12 16 20

SMX recovery (%)"

1D255.75

10225.5

23270.25

1D287.5

1D259. 5

2D301.75

105 105 104 104 96 96 93 90 87

92 98 102 104 97 104 107 103 106

108 103 105 109 ----

101 101 101 101 101 101 100 100 101

112 110 110 104 102 101 100 99 99

98 98 98 98 98 100 97 98 99

a Average of two determinations

a Average for three determinations

at 225.5 nm, and S M X gave straight lines up to 25 mg/1 at 259.7 nm and up to 35 mg/1 at 287.5 nm. In the second derivative calibration graphs were obtained between 1 and 35 mg/1 for T M P and SMX. Table 4 shows the results obtained in the determination o f S M X and T M P in synthetic mixtures. S M X gave g o o d values of recovery by the first derivative at 287.5 nm, and by the second derivative at 301.75 nm, whereas T M P gave the best results by the first derivative at 225.5 rim. In the second derivative, T M P could be determined only for mixtures from 2:1 to 5 : 1 ( T M P : SMX). Table 5 summarizes the recoveries obtained for S M X and T M P in pharmaceutical preparations. S M X gave recovery values between 104% and 102% by the first derivative at 287.5 nm and between 91% a n d 102% at 259.5 nm. In the second derivative the values were between 99% and 104%. The best recoveries o f T M P for all o f the three c o m p o u n d s were obtained by the first derivative at 225.5 nm, whereas by the second derivative the three c o m p o u n d s could not be determined. W i t h the d a t a obtained, we can c o m p a r e the use of the ratio spectra derivative and the classical derivative spectrop h o t o m e t r y for resolving mixtures of T M P and SMX. Statistical parameters show that b o t h methodologies are very similar; however, for the resolution of synthetic mixtures in pharmaceutical p r e p a r a t i o n s the ratio spectra derivative gave the best results with recovery values between 97% and 104%.

Comparative study

Conclusions

F o r m a k i n g a comparative study, the same synthetic and pharmaceutical mixtures were resolved by using the classical zero-crossing measurements [16] in the first and second derivative spectra. Figure 5 shows the first derivative spectra o f SMX, T M P and their mixture. By first derivative m o d e S M X can be determined by measuring at 287.5 n m (1D28v.s) and 259.5 nm (1D259.5) , and T M P can be determined at wavelengths 255.75 nm (1D255.75) and 225.5 n m (tD225.5). In the second derivative mode, S M X can be determined at 301.75 n m (2D301.75) and T M P at 270.25 n m (2D270.25). Some values o f A2 were tested for obtaining the first and second derivative spectra and 15 nm and 25 nm were considered, respectively, as the o p t i m u m values. C a l i b r a t i o n graphs were obtained at the wavelengths selected before (Table 1). In the first derivative, T M P gave straight lines up to 50 mg/1 at 255.75 n m and up to 15 rag/1

These results show that the m e t h o d p r o p o s e d is effective and suitable for the simultaneous determination o f S M X and T M P in pharmaceuticals with better accuracy than classical derivative spectrophotometry.

Table 5. Results obtained in the determination of TMP and SMX

in commercial pharmaceutical preparations by using the proposed method and the zero-crossing method Recovery (%)" Mucorama 80 mg TMP, 400 mg SMX and 120 mg iodopropylidenglycol per tablet

Eduprim 80 mg TMP and 400 mg SMX per tablet

Septrim 80 mg TMP and 400 mg SMX per tablet

101 102

101 100

TMP 1DDz44

97

1DD293

---

1DD244, 293

100

100

1D225.5

71 108

82 88

104 94

2D27o.25

--

__

__

100 99 99 104 91 100

99 99 99 101 96 99

101 101 101 102 102 104

1D255.75

SMX 1DD252.5 1DD264 1DD252.5,264 1D287. 5

1D259.5 2D301. 7

Acknowledgement.We thank

the D.G.I.C.Y.T. of the Ministerio de Educacion y Ciencia (Spain) for supporting this study. Project PB 90-0397.

References

1. Thomas MH, Epstein RL, Ashworth RB, Marks H (1983) Assoc Off Anal Chem 66 : 8 8 4 - 892 2. Stout SJ, Steller WA, Manuel A J, Poeppel MO, da Cunha AR (1984) J Assoc Off Anal Chem 67:142-144

728 3. Simpson RM, Suhre FB, Schafer JW (1985) J Assoc Off Anal Chem 68:23 - 2 6 4. Haagsma N, van de Water C (1985) J Chromatog 333:256261 5. Aerts MML, Beekand WMJ, Brinkman UATh (1988) J Chromatog 435:97-112 6. Bratton AC, Marshall EK (1939) J Biol Chem 128:537-550 7. Bye A, Fox AFJ (1974) Clin Chem 20:288-293 8. Koupparias MA, Anagnostopoulou PI (1988) Anal Chim Acta 204:271 - 283 9. Salinas F, Espinosa Mansilla A, Berzas Nevado JJ (1990) Anal Chim Acta 233 : 2 8 9 - 294

10. Salinas F, Berzas Nevado JJ, Espinosa Mansilla A (1990) Talanta 37: 3 4 7 - 351 11. Berzas Nevado JJ, Guiberteau Cabanillas C, Salinas Lopez F (1990) Anal Lett 23 : 2077- 2094 12. Berzas Nevado JJ, Guiberteau Cabanillas C, Salinas Lopez F (1991) Bull Soc Chim Bel 100:79-91 13. Berzas Nevado J J, Rodriguez Flores J, de la Morena Pardo ML (1991) Talanta (in press) 14. Beckman Instruments (1987) Spectroscopy 2:16 15. Savitzky A, Golay MJE (1964) Anal Chem 36 : 1627-1639 16. O'Haver TC, Green GL (1976) Anal Chem 48:312-318