98 Acta Chim. Slov. 2005, 52, 98–103 Technical Paper Simultaneous Determination of Salicylamide and Paracetamol by Spectrophotometric H-Point Standard Addition Method and Partial Least Squares Regression Abbas Afkhami* and Nahid Sarlak Department of Chemistry, Faculty of Sciences, Bu-Ali Sina University Hamadan, Iran E-mail: afkhami@basu.ac.ir Received 14-03-2004 Abstract Simultaneous spectrophotometric determination of salicylamide and paracetamol by H-point standard addition method (HPSAM) and partial least squares (PLS) calibration is described. The results showed that simultaneous determinations could be performed with the ratio 0.2:5-20:1 for salicylamide - paracetamol. A partial least - squares multivariate calibration method for the analysis of binary mixtures of paracetamol and salicylamide was also developed. The total relative standard error for applying the PTS method to 10 synthetic samples in the concentration range 0-60 /xg mL4 salicylamide and 0-30 /xg mL4 paracetamol was 5.1%. Both the proposed methods (PTS and HPSAM) were successfully applied to the determination of salicylamide and paracetamol in pharmaceutical preparations. Key words: Salicylamide, Paracetamol, Simultaneous determination, HPSAM, PTS Introduction Acetaminophen (/V-acetyl-p-aminophenol; paracetamol) and salicylamide (o-hydroxybenzamide) have been widely used as analgesic and antipyretic drugs. They are frequently prescribed in admixture with each other or with other related drugs. Several methods have been reported for simultaneous determination of salicylamide and paracetamol. These include HPLC,1 spectrofluorimetric,2 electrochemical,3 and spectrophotometric4 methods. The H-point standard addition method5 (HPSAM) permits both proportional and constant errors produced by the matrix of the sample to be corrected directly. This method was based on the principle of dual wavelength spectrophotometry and the standard addition method. The greatest advantage of HPSAM is that, it can remove the errors resulting from the presence of an interfering and blank reagent. Although HPSAM could remove the error resulting from the sample matrix, it cannot remove the constant error resulting from other components in the system. The requirements for the application of the method is that if necessary to work only at two wavelengths where the analvtical signal due to the one of the species is constant and for another one to be as different as possible. By plotting the analytical signal versus the added analyte concentration, two straight lines are obtained that have a common point with coordinates H (-CH, AH), where -Cg is the unknown analyte concentration and AH the analytical signal due to the interfering species. In recent years, multivariate calibration methods have an increasing importance in multicomeponent analysis, especially those using the (PLS) method with decomposition in to latent variables. Interest in UV-VIS spectrophotometric methods has increased and been renewed through the use of signal processing and multivariate techniques6 such as partial least squares (PLS) regression7'8 and artificial neural netvvorks.9 These tools can allow simultaneous spectrophotometric determination of several elements and drugs as well as improve the data handling process of complex chemical systems. In this work HPSAM and PLS were employed for the resolution of binary mktures of paracetamol and salicylamide. The suggested methods were success-fully applied to the determination of these analytes in pharmaceuticals. Results and discussion Figure 1 shows the absorption spectra of salicyla-mide and acetaminophen. As can be seen, the spectra of both the compounds show a strong overlapping hin-dering the resolution of their mixture by conventional spectrophotometry. However, the system is suitable for the simultaneous determination salicylamide and paracetamol using HPSAM. In order to find the optimum pH for the determination of salicylamide and paracetamol by spectrophotometric method, the influence of pH in the range 2-9 on their spectra was investigated. The results showed that pH in the range 2-9 had no effect on the determina- Afkhami and Sarlak Determination of Salicylamide and Paracetamol Acta Chim. Slov. 2005, 52, 98–103 99 Table 1. Characteristics of calibration graphs for the determination of salicylamide and paracetamol by the proposed method. Analyte Salicylamide Paracetamol ~ Slope Intercept Correlation Coefficient 0.0267 0.0659 0.02 0.174 0.999 0.9987 Linearrange/ngmL-1 0.20-60 0.50-30 Limit of Detection/ug mL'1 0.09 0.31 tion of salicylamide and paracetamol. Therefore routine works were performed at pH 6. Individual calibrations To verify the governing Beer’s law, calibration graphs were prepared for the determination of salicy-lamide and paracetamol at pH 6. Characteristics of the calibration graphs are given in Table 1. The Limit of detection was defined as CL = 3SB/m, where CL, SB and m are limit of detection, standard deviation of the blank signal and slope of the calibration graph, respectively.10 The appropriate correlation coefficients obtained indicate that the interaction betvveen the two binary systems either does not exist or at least does not affect the linear correlation prevailing betvveen absorbance and concentration of each drug. Thus, chemometric methods based on factor analysis such as PLS seem to be suitable for use in this system. 3 2.5 2 1.5 1 0.5 230 280 330 380 Wavelength/nm Figure 1. Absorption spectra of 20 fig mL-1 salicylamide, (a) 20 fig mL-1 paracetamol (b) and their mixture (c) at pH 6. Requirements for applying HPSAM Consider an unknown sample containing an analyte X and an interferent Y. The determination of concentration of X by HPSAM under these conditions requires the selection of two wavelengths ^ and X2 at which the interfering species, Y, should have the same absorbance. Then known amounts of X are successively added to the mixture and the resulting absorbances are measured at the two wavelengths and expressed by equations (1) and (2), where Anr, and An2) are the analytical signals measured at ^ and X2, respectively; b0 and Aq (b0^ Aq) are the original analytical signal of X at Xj and X2 respectively; b and A are the analytical signals of Y at \ and X2, respectively; Mu and Ml2 are the slopes of the standard addition calibration lines at ^ and X2, respectively; Q is the added X concentration. The two straight lines obtained intersect at the so-called H point (-CH, AH), (Figure 2, equations (1) and (2)). At the H-point, since A(U)= A„2), Q = -Cjj from equations (1) and (2) follow equations (3) and (4). 2.5 - 2 1.5 1 0.5 -5 cH 5 15 25 Caddedsalicylamide/ngmL-1 35 Figure 2. Plots of H-point standard addition method for fixed salicylamide (2 fig mL-1) and (¦) 2 and (A) 5 fig mL-1 of paracetamol. From Equation (4), the following conclusions can be obtained: (i) If the Y component is the known interferent and the analytical signal corresponding to Y, b (at Xj) and A (at X2) do not change with the additions of an analyte, X, that is, b = A = constant, and thus see equations (5)-(8). A(U) = b0 + b + Mu Q, (1) A(X2) = A0 + A' + MX2 Q, (2) b0 + b + MU(-CH) = A0 + A' + MX2(-CU), (3) _CH = [(A0 - b0) + (A' - b)] / (Mu - MX2). (4) Cx = (A0 - b0)/MXl - MX2) = b0/MXl = Ao/MX2, (5) If CH= - Cx then, _CH= (Ao-b0) / (Mu - MX2) = b,/ Mu = Ao/+ MX2 (6) If the value of -CH is included in Equation (1), then AH= b0+ b+ Mu (-CH) (7) B0= - MiA CH (Equation (4)), then AH= b (8) And similarly 0 AH= A'. Afkhami and Sarlak Determination of Salicylamide and Paracetamol 100 Acta Chim. Slov. 2005, 52, 98–103 Table 2. Results of several experiments for the analysis of salicylamide-acetaminophen mktures at different concentration ratios by HPSAM. A- C Equation Present in sample / ng mL1 Found / ng mL"1 Salicylamide Paracetamol Salicylamide Paracetamol A233 = 0.3425 + 0.054Ci 0.999 0.20 5.00 A252= 0.3345 + 0.014Ci 0.996 A233 = 0.3088 + 0.0536C; 0.999 3.00 2.00 A252= 0.1878 + 0.0130Ci 0.999 A233= 0.439 + 0. 058Cj 0.998 5.00 2.00 A252= 0.213+ 0.01216Ci 0999 A233= 0.272 + 0.0545Ci 0.999 2.00 2.00 A252= 0.190 + 0.0133Ci 0.997 A233= 0.4283 + 0.055Ci 0.999 2.00 5.00 A252= 0.3494 + 0.013Ci 0.998 A233= 0.3987 + 0.05546Ci 0.999 1.00 5.00 A252= 0.3591+ 0.01286Ci 0.996 A233= 0.2654 + 0.054Ci 0.999 1.00 3.00 A252= 0.2228 + 0. 0131Ci 0.999 A233= 0.987 + 0.051Ci 0.999 17.00 1.00 A252= 0.304 + 0.012Q 0.998 A233= 1.199 + 0.05Ci 0.999 20.00 1.00 A252= 0.372 + 0.00979Ci 0.998 0.20 2.98 4.93 1.99 1.88 0.93 1.04 17.50 20.57 4.96 2.06 2.10 2.02 4.75 5.10 2.99 1.00 1.00 Hence, AH value is only related to the signal of the interference Y at the two selected wavelengths and Cjj is independent from the concentration of interference. Figure 2 shows the effect of change in concentration of paracetamol on the position of the H-point. (ii) If component Y is the unknown interferent, Equation (4) is tenable as long as the Y analytical signals (b at Xj and A’ at X2) remain equal with the addition of analyte X. According to the above discussion at H point, CH is independent from the concentration of interferent and so, AH is also independent from the analyte concentration. For selection of appropriate wavelengths for ap-plying HPSAM, the following principles were followed. At two selected wavelengths, the analyte signals must be linear with the concentrations, the interferent signals must remain equal, even if the analytical concentrations are changed, and the analytical signals of the mkture composed from the analyte and the interferent should be equal to the sum of the individual signals of the two compounds. In addition, the slope difference of the two straight lines obtained at ^ and X2 must be as large as possible in order to get good accuracy. In this special system, analyte is salicylamide and paracetamol is as interferent. Several wavelength pairs were examined and the wavelength pair of 233 and 252 nm was selected. Under optimum conditions, deter-mination of salicylamide and paracetamol was carried out using HPSAM. The concentration of interferent was calculated in each test solution by the calibration method with a single standard and the ordinate value of the H- point (AH). Several synthetic mktures with different concentration ratios of salicylamide and paracetamol were analyzed by the proposed method. The results are given in Table 2. Repeatability of the HPSAM Simultaneous determination of salicylamide and paracetamol at pH 6.0 were made using HPSAM. To check the repeatability of the method five replicate experiments of the salicylamide and paracetamol were done (Table 3). Then the concentration of interferent was calculated in each test solution by calibration method using standard solutions and the ordinate value of H- point (AH). The relative standard deviations (RSD) for five replicate measurements of the mkture of 2.0 (ig mL4 each of paracetamol and salicylamide werel.5 and 2.48%, respectivelv. Partial least squares (PLS) regression The calibration procedure consisted of a complete experimental design with six concentration levels for both paracetamol and salicylamide. Each solution was prepared to contain combina-tions of the concentration levels (0.0-60 jig mL4 of salicylamide and 0.0-30 jig mL4 of paracetamol). A set of 33 mktures of paracetamol and salicylamide are shown in Table 4. R Afkhami and Sarlak Determination of Salicylamide and Paracetamol Acta Chim. Slov. 2005, 52, 98–103 101 Table 3. Results for five replicate for the analysis of paracetamol - salicylamide mixtures by HPSAM. Found/ngmL-1 A- C Equation R Presentinsample/ngmL-1 Salicylamide Paracetamol Salicylamide Paracetamol A233 0.297+ 0.053C; 0.999 2.00 2.00 1.94 2.06 A252= 0.220+0.0133Ci 0.999 A233 = 0.272+ 0.0545Ci 0.999 2.00 2.00 1.99 2.02 A252= 0.190+ 0.0133Ci 0.998 A233= 0.241 + 0. 055Ci 0.999 2.00 2.00 2.02 2.04 A252= 0.156+0.013Ci 0.996 A233= 0.278 + 0. 053Cj 0.999 2.00 2.00 2.00 1.93 A252= 0.195+ 0.0118Ci 0.996 A233= 0.275 + 0.0557Ci 0.999 2.00 2.00 2.00 2.00 A252= 0.189+0.0127Ci 0.999 Mean 2.01 1.99 Standard deviation 0.05 0.03 R.S.D (%) 2.48 1.50 Table 4. Values of the paracetamol and salicylamide concentra-tions used as calibration and prediction solutions in fig mL-1. Calibration set Prediction set Paracetamol Salicylamide Salicylamide Paracetamol 0 12 36 0 0 24 12 6 0 48 60 6 0 60 48 12 6 0 32 18 6 12 60 18 6 36 0 24 6 48 36 24 12 0 12 30 12 12 48 30 12 24 - - 12 36 - - 12 60 - - 18 0 - - 18 12 - - 18 36 - - 18 48 - - 24 12 - - 24 24 - - 24 48 - - 24 60 - - 30 0 - - 30 24 - - Twenty-three of these solutions were used as a calibration set for PLS model development. Another 10 calibration mixtures, not included in the previous set were employed as an independent test set called the prediction set. To select the number of factors in the PLS algorithm, the cross- validation method, employed was to eliminate only one sample at a tirne11. The prediction error was calculated for each component for the prediction set, which are the samples not participating in the construction of the model. The optimum number of factors (latent variables) to be included in the calibration model was determined by computing the prediction error sum of squares (PRESS) for the first variable, which built the PLS modeling in the calibration step, then, another latent variable was added for the model building and the PRESS was calculated again. This proc-ess was repeated for one to 10 latent variables, which were used in the PLS modeling. The predicted concentrations of the compounds in each sample were compared with the already known concentration and the prediction error sum of squares (PRESS) was calculated by each number of factors. Figure 3 shows a plot of PRESS against the number of factors for each individual component. The F-statistical test can be used to determine the significance of PRESS values greater than the minimum. The optimal number of factors for paracetamel and salicylamide was ob-tained too. The results obtained are given in Table 5. The prediction error of a single component in the mixture was calculated as the relative standard error (R.S.E) of the prediction concentration (Equation (9))1213 where N is the number of samples, Cj the concentration of the component in the jth mkture and Čj the estimated concentration. The total prediction error of N samples is calculated as follows in Equation (10), where C;j is the concentration of the ith component in the jth sample and Č;j its estimation. Table 5 also shows reasonable single and total relative errors for such a system. Afkhami and Sarlak Determination of Salicylamide and Paracetamol 102 Acta Chim. Slov. 2005, 52, 98–103 Table 5. Composition of prediction set, their predictions by PLS model and statistical parameters for the system. Composition (ng mL4) Prediction (ng mL"1) Recovery (%) Salicylamide Paracetamel Salicylamide Paracetamal Salicylamide Paracetamal 36.00 0.00 36.00 0.09 100.0 - 12.00 6.00 12.50 6.71 104.1 111.7 60.00 6.00 60.97 6.43 101.6 107.1 48.00 12.00 48.00 12.55 100.0 104.6 32.00 18.00 35.42 19.60 110.7 108.9 60.00 18.00 60.56 18.69 100.9 103.8 0.00 24.00 0.48 25.46 - 106.0 36.00 24.00 36.27 25.08 100.7 104.5 12.00 30.00 11.74 34.86 97.8 116.2 48.00 30.00 45.46 31.79 94.7 105.9 Mean recovery 101.1 108.6 R.S.E.(%) single 3.57 8.7 Table 6. Simultaneous determination of salicylamide and paracetamol in pharmaceutical preparations by HPSAM and PLS method. Sample" Nominalvalue/mg Found*/mg Salicylamide Paracetamol Salicylamide Paracetamol Yendolgranularpacket 500 200 510 ±5" 493 ±6" 207 ± 3C 201 ±4" Pridio capsules 100 300 104 ± 2C 96 ±4d 299 ± (f 299 ±3" Rinomicinc activated tablets 150 150 151 ±5C 146±6rf 155 ± T 157±5rf Rinomicinc pellets 50 50 50±3C 49±4rf 50±3C 49±5rf "Composition of samples: Yendol: Salicylamide, 500 mg; paracetamol. 200 mg; chlorpheniramine maleate, 3 mg; caffeine, 39 mg; saccharin, 10 mg; saccharose. 6.56 g. pridio: salicylamide, 100 mg; paracetamol, 300mg; caffeine, 25 mg. chlorpheniramine maleate, 2mg. Rinomicine activated: paracetamol, 150 mg; chlorpheniramine maleate, 4 mg; caffeine, 30 mg; Salicylamide, 150 mg; phenlephrine hydrochloride. 10 mg; Rinomicine: paracetamol, 50mg; chlorpheniramine maleate, 4 mg; Salicylamide, 50 mg; phenlephrine hydrochloride. 20 mg.h Mean ± S. D. (n=3). c By HPSAM. rfBy PLS method. R.S.E.(%) = ( 2i_iM (Č-C^f^.iiCff^ 100 (9) R.S.E., (%) = (S.,MS.,N(QJ-C,J)2^,MSN.,(C1J)2)0-5X100 (10) 3500 - \ a 3000 - 2500 2000 1500 - b \ 1000 500 0 - T-'B— ----D---- i—n— 12 3 4 5 numberoffactors Figure 3. Plot of PRESS against the number of factors for (a) paracetamol and (b) salicylamide. Application To evaluate the analytical applicabilitv of both the proposed methods, (PLS and HPSAM) they were ap-plied to the simultaneous determination of salicylamide and paracetamol in pharmaceutical preparations con-taining both compounds. The results are given in Table 6. The good agreement betvveen the results with the composition values indicated by the suppliers indicates the successful applicability of the proposed methods for simultaneous determination of salicylamide and paracetamol in pharmaceutical preparations. Conclusion The above results show that HPSAM and PLS regression allow rapid, accurate and simple resolution of paracetamol and salicylamide mktures. The HPSAM can be used in the complex samples with matrix effects because standard addition method Afkhami and Sarlak Determination of Salicylamide and Paracetamol Acta Chim. Slov. 2005, 52, 98–103 103 has capability of removing these effects. But partial least squares regression cannot be used in these cases. On the other hand the PLS method was more rapid than HPSAM. Therefore in the mktures with matrix effects HPSAM is preferred but in the mktures without these effect PLS is better than HPSAM because of rapidity. Experimental Reagents Triply distilled water and analytical-reagent grade chemicals were used. A 1000 jig mL4 salicyla-mide (Aldrich) solution was prepared in 5% ethanol (Merck)/ water (v/v); this solution was stable for at least two weeks. Working solutions were prepared daily by diluting the standard solution with water. Standard pa-racetamol solution, 1000 jig mL4 of 4-acetamidophenol (Merck) in water was prepared daily. Working solutions were prepared by diluting the standard solution with water. Apparatus A Pharmacia model LKB3 UV-Visible Ultraspect3 single beam spectrophotometer with 1-cm quartz cells, connected to a Pentium II computer was used for ab-sorbance measurements. AH spectral measurements were performed using the blank solution as a reference. Measurements of pH were made with a Jenway C15 pH- meter using a combined glass electrode. The computations were made with a Pentium4 computer. Ali programs in the computing process were written in MATLAB for windows. Procedure A 1 mL of pH 6 buffer solution and appropriate volumes of salicylamide and acetaminophen solutions were added to a 10 mL volumetric flask and made up to the mark with water and mixed well. Simultaneous determination of salicylamide and acetaminophen with HPSAM was performed by measuring the absorbance of the solution at 233 and 252 nm for each sample. Syn- thetic samples containing different concentration ratios of salicylamide and acetaminophen were prepared and standard addition of salicylamide were made. The concentration range of salicylamide and acetaminophen for the construction of HPSAM calibration graphs were 0.2-30 and 0.5-30 jig mL4, respectively. Simultaneous determination of salicylamide and acetaminophen with PLS method was performed by recording the absorbance spectra for each solution from 230 to 330 nm. The concentration range of sali-cylamide and paracetamol in the PLS method in the same conditions was 0.0-60 jig mL4 and 0.0-30 jig mL4, respectively. References 1. M. E. EI- Kommos. K. M Emara, Talanta 1989, 36, 678-679. 2. K. W. Jim Street, G. H. Schenk,/. Pharm. Sci. 1981, 70, 641-645. 3. M. I. Walash, A. M. El-Brashy, M. A. Sultan. Mikrochim. Acta 1994, 113, 113-124. 4. A. R. Medina, M. L. Fernández de Córdova, A. Molina T)i&z, Anal. Chim. Acta 1999, 394, 149-158. 5. E Bosch-Reig, P. Compins- Falco, Analyst 1988, 113, 1011-1016. 6. R. Lobinski, Z. Marczenko, Crit. Rev. Anal. Chem. 1992, 23, 55-111. 7. H. Martens, T. Naes, Multivariate Calibration, Wiley, Chichester, 1991. 8. M. J. Ayora Canada, M. I. Pascual Reguera, A. Molina Dfaz, L. E Capitán-vallvey, Talanta 1999, 49, 691-701. 9. J. Zupan, J. Gasteiger, Neural Netvvorks for Chemists: An Introduction, VCH. Weinhiem, 1993. 10. D. Perez-Bendito, M. Silva, Kmetic Methods in Analytical Chemistry, Ellis Honvood, Chichester, 1988. 11. D. M. Haaland, E. V Thomas, Anal. Chem. 1988, 60, 1193-1202. 12. M. Otto, W. Wegscheider,^4wa/. Chem. 1985, 57, 63-69. 13. M. Blanco, Coello, H. Ituriaga, S. Maspoch, M. Redon, Appl. Spectroc. 1994, 48, 37-42. Povzetek Opisano je sočasno spektrofotometrično določevanje salicilamida in paracetamola z metodo standardnega dodatka pri dveh valovnih dolžinah (HPSAM) in umeritvijo na osnovi parcialne regresijske analize (PLS). Rezultati so pokazali, da je sočasno določevanje obeh spojin možno v območju razmerij salicilamid:paracetamol od 0,2:5 do 20:1. Skupna relativna standardna napaka pri analizi 10 sintetičnih vzorcev binarnih zmesi s koncentracijami salicilamida 0 – 60 µg mL–1 in paracetamola 0 – 30 µg mL–1 z metodo PLS je bila 5,1 %. Obe metodi (PLS in HPSAM) sta bili uspešno uporabljeni za določevanje salicilamida in paracetamola v farmacevtskih pripravkih. Afkhami and Sarlak Determination of Salicylamide and Paracetamol Acta Chim. Slov. 2005, 52