Validated spectrophotometric methods for simultaneous determination of oxytetracycline associated with diclofenac sodium or with piroxicam in veterinary pharmaceutical dosage form

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 Contents lists available at ScienceDirect Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa Validated spectrophotometric methods for simultaneous determination of troxerutin and carbazochrome in dosage form Fatma I. Khattab a, Nesrin K. Ramadan a, Maha A. Hegazy a,⇑, Medhat A. Al-Ghobashy a,b, Nermine S. Ghoniem a a b Analytical Chemistry Department, Faculty of Pharmacy, Cairo University, Egypt Biotechnology Center, Faculty of Pharmacy, Cairo University, Egypt h i g h l i g h t s g r a p h i c a l a b s t r a c t  Troxerutin (TXN) is co-formulated with Carbazochrome (CZM).  No method was reported for determination of TXN and CZM in their mixture form.  Four spectrophotometric methods were developed for their simultaneous determination.  The methods were validated according to the ICH guidelines.  The results were statistically compared to the manufacturer’s method. a r t i c l e i n f o a b s t r a c t Four simple, accurate, sensitive and precise spectrophotometric methods were developed and validated for simultaneous determination of Troxerutin (TXN) and Carbazochrome (CZM) in their bulk powders, laboratory prepared mixtures and pharmaceutical dosage forms. Method A is first derivative spectrophotometry (D1) where TXN and CZM were determined at 294 and 483.5 nm, respectively. Method B is first derivative of ratio spectra (DD1) where the peak amplitude at 248 for TXN and 439 nm for CZM were used for their determination. Method C is ratio subtraction (RS); in which TXN was determined at its kmax (352 nm) in the presence of CZM which was determined by D1 at 483.5 nm. While, method D is mean centering of the ratio spectra (MCR) in which the mean centered values at 300 nm and 340.0 nm were used for the two drugs in a respective order. The two compounds were simultaneously determined in the concentration ranges of 5.00–50.00 lg mL1 and 0.5–10.0 lg mL1 for TXN and CZM, respectively. The methods were validated according to the ICH guidelines and the results were statistically compared to the manufacturer’s method. Ó 2014 Published by Elsevier B.V. Article history: Received 14 August 2014 Received in revised form 1 December 2014 Accepted 16 December 2014 Available online 24 December 2014 Keywords: Carbazochrome Troxerutin Ratio subtraction Derivative Mean centering Spectrophotometry Introduction Troxerutin (TXN) is chemically designated as 2-[3,4-bis(2hydroxyethoxy)phenyl]-5-hydroxy-7-(2-hydroxyethoxy)-4-oxo⇑ Corresponding author. Tel.: +20 01112887066. E-mail addresses: mahahgazy@yahoo.com, (M.A. Hegazy). http://dx.doi.org/10.1016/j.saa.2014.12.047 1386-1425/Ó 2014 Published by Elsevier B.V. maha.hegazy@pharma.cu.edu.eg 4H-chromen-3-yl6-o-(6-deoxy-b-D-mannopyranosyl)-b-D-glucopyranoside, it is a flavonol and known as vitamin P4 [1]. TXN has a considerable broad pharmacological activities; it improves capillary function, reduces capillary fragility and abnormal leakage. It is also used for reducing the occurrence of night cramps, treatment of varicose veins and hemorrhoids [2]. Carbazochrome (CZM) is chemically designated as [(3-Hydroxy-1-methyl-6-oxo-2,3dihydroindol-5-ylidene)amino]urea and it is an oxidation product F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 of adrenaline and used as anti-hemorrhagic agent [3]. Both drugs are co-formulated (TXN 150 mg and CZM 1.5 mg) and the dosage form was shown to have a good efficacy and safety profile in non-surgical patients with acute uncomplicated hemorrhoids. Their chemical structures are shown in Fig. 1. Several methods were reported on each drug individually. For TXN, it was determined by spectrophotometry [4–6], capillary electrophoresis [7] and high-performance liquid chromatography (HPLC) [8–10], electrochemistry [11] and capillary electrochromatography [12]. While for CZM, it was determined by spectrophotometry [13], high-performance liquid chromatography (HPLC) [14,15] and Chemiluminescence [16]. No method was reported for determination of TXN and CZM in their mixture form. So, our aim was to develop and validate simple spectrophotometric methods for simultaneous determination of both drugs in their pure form, laboratory prepared binary mixture and pharmaceutical formulation. 207 prepared by an additional dilution of their stock standard solutions with methanol. A set of laboratory prepared mixtures containing different ratios of TXN (10.0–50.0 lg mL1) and CZM (0.5– 7.5 lg mL1) was prepared. Procedures Construction of calibration curves Aliquots of TXN working standard solution (0.1 mg mL1) equivalent to 50.0–500.0 lg mL1 and of CZM working standard solution (0.1 mg mL1) equivalent to 5.0–100.0 lg mL1 were accurately transferred into a series of 10 mL volumetric flasks; the volume was completed to the mark with methanol. The zero order spectra of the prepared solutions were recorded using methanol as a blank in the range of 200–600 nm. For D1 method Experimental Instruments A double beam UV–visible spectrophotometer (SHIMADZU, Kyoto, Japan) model UV-1650, pc with quartz cell of 1 cm path length, connected to IBM compatible computer operated with UV-probe personal spectroscopy software version 2.21. The spectral band width is 2 nm and wavelength scanning speed is 2800 nm/min. Mean centering computations were done using MatlabÒ 6.5 with PLS-Toolbox. The D1 curves of the scanned spectra were recorded using Dk = 4 and scaling factor = 10. Calibration curves were then constructed by plotting the values of the peak amplitude of D1 curves at 294 nm for TXN (corresponding to zero crossing of CZM) and 483 nm for CZM (corresponding to zero absorbance of TXN) versus the corresponding concentrations and the regression parameters were computed. For DD1 method Pure samples of TXN and CZM were kindly supplied by Minapharm for Pharmaceuticals and Chemical Industries, Cairo, Egypt. Both were certified to contain 99.90% w/w according to the manufacturer’s method. FlebotonÒ ampoules, labeled to contain 150 mg of TXN and 1.5 mg of CZM were manufactured by Minapharm Pharmaceuticals and Chemical Industries, Egypt (batch No. 9DE0219) and were obtained from local market. The scanned spectra of TXN were divided by a standard spectrum of 10.0 lg mL1 CZM while the spectra of CZM were divided by a standard spectrum of 40.0 lg mL1 TXN and the first derivative of the ratio curves (DD1) for each compound were then obtained with Dk = 4 and scaling factor = 10. Calibration curves were constructed by plotting the peak amplitude at 248 and 439 nm of the DD1 curves versus the corresponding concentrations of TXN and CZM, respectively and the regression parameters were computed. Chemicals and solutions For ratio subtraction method (RS) Methanol spectroscopic grade was used. Stock standard solutions of TXN and CZM (1.0 mg mL1) were prepared in methanol. Working standard solutions of TXN and CZM (0.1 mg mL1) were For the determination of TXN, a calibration curve was constructed relating the absorbance of zero order spectra of TXN at 352 nm to the corresponding concentrations and the regression Samples Fig. 1. Chemical structure of (a) truxerutin and (b) carbazochrome. 208 F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 equation was computed while CZM is determined as previously mentioned in D1. For MCR method The stored spectra of the prepared solutions of TXN were divided by the normalized absorption spectrum of CZM. Similarly, the zero-order spectra of the prepared solutions of CZM were divided by the normalized absorption spectrum of TXN. The obtained ratio spectra were then mean centered. The calibration curves for the two drugs were constructed by plotting the mean centered values at 300 nm and 340 nm versus the corresponding concentrations of TXN and CZM, respectively. Application to laboratory prepared mixtures The absorption spectra of the laboratory prepared mixtures were recorded. Then the procedures were followed as described under construction of the calibration curves. The concentrations of TXN and CZM were calculated by substituting in the corresponding regression equation. D1 method It is a very useful analytical technique for eliminating spectral overlapping by using the first or higher derivatives of absorbance with respect to wavelength [17]. Upon applying D1 method, TXN could be determined by measuring its peak amplitude at 294 nm (corresponding to zero-crossing of CZM), while CZM could be determined by measuring its peak amplitude at 483 nm (corresponding to zeroabsorbance of TXN) (Fig. 3). In order to optimize D1 method, different smoothing and scaling factors were tested, where a Dk = 4 and a scaling factor = 10 showed a suitable signal to-noise ratio and the curves showed good resolution. A linear correlation was obtained between the peak amplitude and its corresponding concentration for TXN at k = 294 nm and for CZM at 483 nm in the ranges of 5.0–50.0 lg mL1 and 0.5–10.0 lg mL1 for TXN and CZM, respectively. The parameters of the regression equations are shown in Table 1. DD1 spectrophotometric method Three Fleboton Ampoules were mixed and an accurate volume equivalent to 50 mg of TXN and 0.5 mg of CZM was transferred into a100 mL volumetric flask. Dilution of active ingredients was carried out by addition of methanol. Suitable dilutions were made using methanol to prepare solution containing 50.0 lg mL1 TXN and 0.5 lg mL1 CZM. In order to improve the selectivity of the analysis of TXN and CZM, DD1 was also applied and validated. The main advantage of this method is that the whole spectrum of the interfering substance is canceled [18]. In order to optimize the DD1 method, several divisors were tested along with the normalized spectrum. The best results were obtained using each of 10.0 lg mL1 of CZM and 40.0 lg mL1 of TXN as a divisor for TXN and CZM, respectively. Peak amplitude at 248 nm and 439 nm for TXN and CZM, respectively (Figs. 4 and 5) were selected, plotted against the corresponding concentration. Good linearity was obtained for both drugs and the regression parameters were calculated and shown in Table 1. Results and discussion Ratio subtraction method Truxerutin is co-formulated with Carbazochrome in ampoules for treatment of chronic venous insufficiency. By reviewing the literature in hand, there was no reported spectrophotometric method for their simultaneous determination. The absorption spectra of TXN and CZM show severe overlap (Fig. 2) which hinders their direct spectrophotometric determination. So, our aim was to develop simple, rapid, accurate and precise spectrophotometric methods for the simultaneous determination of TXN and CZM in dosage form. The method was applied for determination of mixture of TXN (X) and CZM (Y), when the spectrum of (Y) extended than the other (X), as shown in Fig. 2. The determination of (X) could be achieved by scanning the absorption spectra of the laboratory prepared mixtures of X and Y in methanol, then dividing them by a carefully chosen standard spectrum of Y (10.0 lg ml1, Y0 = divisor) to produce a new ratio spectra that represents X/Y0 + constant, as shown in Fig. 6. The values of these constant (Y/Y0 ) were subtracted as shown in (Fig. 7), followed by multiplication of the obtained spectra by the divisor (Y0 ) as shown in (Fig. 8). Finally, the original spectra of X were obtained and the absorbance values at 352 nm were plotted against the concentration, the corresponding regression equation could be calculated. This can be summarized as follows: Application to pharmaceutical formulation xþ y x y x ¼ þ ¼ þ constant y0 y0 y0 y0 x x þ constant  constant ¼ 0 y0 y x  y0 ¼ x y0 Fig. 2. D0 absorption spectra of 1.25 lg mL1 of CZM, and 50 lg mL1 of TXN in methanol. The constant can be determined directly from the curve (X + Y)/Y0 by the straight line which is parallel to the wavelength axis in the region where (Y) is extended. The correct choice of the divisor is a fundamental step, as if the concentration of the divisor increases or decreases, the resulting constant value will be proportionally decreased or increased [18].A linear correlation was obtained between the absorbance and the corresponding concentration of TXN at its corresponding wavelength, the parameters of the regression equations are shown in Table 1. While, CZM was previously determined by D1 method. 209 F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 Fig. 3. D1 of 10 lg mL1 of CZM, and 50 lg mL1 TXN in methanol. Table 1 Validation and regression parameters obtained by applying the proposed methods. Parameter Range Slope Intercept SE of the slope SE of the intercept Correlation coefficient (r) LOD LOQ Accuracy (mean ± SD) RSD% D1 DD1 Ratio subtraction MCR CZM TXN CZM TXN CZM TXN CZM TXN 0.5–10 lg mL1 0.0492 0.0048 0.305  103 1.590  103 0.9999 0.167 0.508 100.79 ± 0.538 0.534 5–50 lg mL1 0.0128 0.0117 0.086  103 2.855  103 0.9999 0.762 2.308 99.74 ± 0.510 0.511 0.5–10 lg mL1 0.3815 0.0822 1.839  103 0.012 0.9999 0.174 0.526 99.87 ± 0.715 0.716 5–50 lg mL1 0.0251 0.0128 0.208  103 6.918  103 0.9998 0.956 2.897 99.80 ± 0.894 0.896 0.5–10 lg mL1 0.0492 0.0048 0.305  103 1.590  103 0.9999 0.167 0.508 100.79 ± 0.538 0.534 5–50 lg mL1 0.0263 0.006 0.0002 0.007 0.9998 0.872 2.642 99.30 ± 0.953 0.959 0.5–10 lg mL1 40.3838 1.4507 0.101 0.525 0.9999 0.068 0.206 99.42 ± 0.567 0.571 5–50 lg mL1 1.7755 0.897 0.018 0.6 0.9998 1.17 3.547 99.45 ± 0.835 0.839 Fig. 4. (a) D0 of TXN (5–50 lg mL1) in methanol, using 10 lg mL1 of CZM as a divisor. (b) DD1 of (5–50 lg mL1) of TXN, using 10 lg mL1 of CZM as a divisor. MCR method For further improvement of the selectivity, a new, simple recently developed method was applied. This is based on the mean centering of ratio spectra. It eliminates the derivative step and so the signal-to-noise ratio is therefore enhanced [19]. The MCR method was applied and was able to quantitatively determine both TXN and CZM in their laboratory-prepared mixtures and in their pharmaceutical preparation. As shown in Fig. 2, the absorption spectra of TXN and CZM in methanol are severely overlapped in the wavelength region of 250–500 nm. So, the absorption spectra of the standard solutions of the TXN with different concentrations were 210 F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 Fig. 5. (a) D0 of CZM (0.5–10 lg mL1) in methanol, using 40 lg mL1 of TXN as a divisor. (b) DD1 of (0.5–10 lg mL1) of CZM, using 40 lg mL1 of TXN as a divisor. Fig. 6. Ratio spectra of laboratory prepared mixtures of CZM and TXN using 10 lg mL1 of CZM as the divisor, in methanol. recorded in the wavelength range of 200–600 nm and divided by the normalized spectrum of the CZM. The ratio spectra were obtained. Mean centering of the ratio spectra was carried out and the concentration of TXN was determined by measuring the amplitude at 300 nm (corresponding to a maximum wavelength) (Fig. 9). The spectra of the standard solutions of the CZM with different concentrations were recorded in the wavelength range of 200– 600 nm and divided by the normalized spectrum of the TXN. The ratio spectra were obtained. Mean centering of the ratio spectra was carried out and the concentration of CZM was determined by measuring the amplitude at 340 nm (corresponding to a maximum wavelength) (Fig. 10). A linear correlation was obtained between the mean centered values and its corresponding concentration for TXN at 300 nm and for CZM at 340 nm. The parameters of the regression equations are shown in Table 1. The effect of divisor concentration on the analytical parameters such as slope, intercept and correlation coefficient of the calibra- tion graphs was also tested. Different divisors were tested; a normalized spectrum of each of TXN and CZM was used as a divisor spectrum in the proposed method. The specificity of the proposed methods was proved by the analysis of laboratory prepared mixtures of TXN and CZM in different ratios, as presented in Table 2. All the proposed methods were successfully applied for the determination of TXN and CZM in FlebotonÒ Amp. (Table 3) and the results obtained were statistically compared with those obtained by the manufacturer method and there is no significant difference regarding both accuracy and precision as shown in Table 4. Conclusion The proposed methods were simple, rapid, sensitive and precise. They could be easily applied in quality-control laboratories for simultaneous determination of TXN and CZM. MCR method F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 Fig. 7. Ratio spectra of laboratory prepared mixtures of CZM and TXN using 10 lg mL1 of CZM as divisor, after subtraction of the constant. Fig. 8. Spectra of laboratory prepared mixtures of CZM and TXN after multiplying the subtracted ratio spectra by the spectra of 10 lg mL1 of CZM. Fig. 9. MCR of ratio spectra of TXN using normalized CZM spectrum as a devisor. 211 212 F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 Fig. 10. MCR of ratio spectra of CZM using normalized TXN spectrum as a divisor. Table 2 Determination of CZM and TXN in laboratory prepared mixtures by the proposed spectrophotometric methods. * ** Mixture no. Concentration (lg mL1) D1 Recovery% DD1 CZM TXN CZM TXN CZM TXN CZM TXN CZM TXN 1 2 3 4 5 6** Mean RSD 7.5 10 5 2.5 1 0.5 10 20 25 30 40 50 98.56 98.35 100.45 100.57 100.2 99.59 99.62 0.970 99.22 98.16 98.72 99.92 101.56 101.55 99.86 1.441 100.2 100.2 98.17 98.63 99.63 98.5 99.21 0.895 100.08 99.44 99.79 99.76 99.22 101.13 99.9 0.672 98.56 98.35 100.45 100.57 100.2 99.59 99.62 0.970 100.57 101.54 100.74 100.77 99.9 101.51 100.84 0.613 99.35 99.57 98.19 99.33 98.55 99.1 99.02 0.539 100.02 98.23 99.65 98.07 98.59 99.56 99.02 0.833 * Ratio subtraction MCR CZM was determined according to its regression equation in D1. Dosage form ratio. Table 3 Statistical analysis of the results of the proposed methods and manufacturer methods for TXN and CZM. Parameter Mean SD N Variance Student’s t F CZM TXN D1 DD1 Ratio subtraction MCR ** manufacturer method D1 DD1 Ratio subtraction MCR ** manufacturer method 100.79 0.538 4 0.289 1.66 (2.31)⁄ 1.43 (5.41)⁄ 99.87 0.714 4 0.509 0.67 (2.31)⁄ 1.23 (9.01)⁄ 100.79 0.538 4 0.289 1.66 (2.31)⁄ 99.42 0.567 4 0.321 1.92 (2.31)⁄ 1.29 (5.41)⁄ 100.17 0.644 6 0.414 99.74 0.510 4 0.260 0.589 (2.26)⁄ 2.84 (4.76)⁄ 99.80 0.893 4 0.797 0.338 (2.26)⁄ 1.08 (8.94)⁄ 99.30 0.953 4 0.908 1.19 (2.26)⁄ 99.45 0.835 4 0.697 1.02 (2.26)⁄ 1.05 (4.76)⁄ 99.99 0.859 7 0.737 1.43 (5.41)⁄ 1.23 (8.94)⁄ The values in parentheses are the corresponding tabulated values at P = 0.05. Two methods were used one for TXN (UV), D0 at 254 nm, and the second for CZM HPLC (using phosphate buffer pH 5.7: Methanol, 55:45 (v/v), C-18 column, flow rate 0.8 mL min1 and detection at 354 nm). * ** Table 4 Application of the proposed methods for the analysis of TXN and CZM in pharmaceutical dosage form. * ** Product D1 DD1 ** Mean recovery% ± R.S.D * FlebotonÒ ampoules 1.5 mg CZM and 150 mg TXN/ampoule B.N.9DE0219 CZM TXN 98.92 ± 0.66 99.58 ± 0.96 CZM was determined according to its regression equation in D1. Average of 3 determination. 100.41 ± 1.18 99.96 ± 0.78 Ratio subtraction MCR 100.41 ± 1.18 99.32 ± 1.00 99.02 ± 0.72 99.01 ± 0.98 F.I. Khattab et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 139 (2015) 206–213 has the advantage of eliminating the derivative steps and therefore the signal-to-noise ratio is not degraded. They methods could be applied for the routine QC analysis in their pure bulk powders and in dosage form without any preliminary separation step. References [1] R. Bruppacher, B. Rieckemann, B. Naser-Hijazi, P. Wüstenberg, Evaluation of the safety of a coumarin–troxerutin combination, Pharmacoepidemiol. Drug Saf. 7 (S1) (1998) S37–S40. [2] M. Basile, S. Gidaro, M. Pacella, P.M. Biffignandi, G.S. Gidaro, Parenteral troxerutin and carbazochrome combination in the treatment of posthemorrhoidectomy status: a randomized, double-blind, placebo-controlled, phase IV study, Curr. Med. Res. Opin. 17–4 (2001) 256–261. [3] M. Verstraete, Report on adrenochrome monosemicarbazide (carbazochrome) and carbazochrome salicylate, in: Haemostatic Drugs, Springer, Netherlands, 1977, pp. 59–66. [4] H. Quan, X.-H. Bai, X. Yang, Determination of troxerutin and trace rutin contents in troxerutin tablets by double wavelength UV spectrophotometry, J. Shanxi Med. Univ. 36–5 (2005) 592. [5] P. Issopoulos, A new spectrophotometric determination of troxerutin using an oxidative coupling reaction of 3-methylbenzothiazolin-2-one hydrazone, Pharmazie 48–1 (1993) 67–68. [6] Y. Suhua, G. Aiqin, J. Zhengping, M. Zhengjun, Z. Xiashi, Determination of troxerutin by spectrophotometry, J. Yangzhou Univ. (Nat. Sci. Ed.) 1 (2006). [7] J. Široká, A. Martincová, M. Pospíšilová, M. Polášek, Assay of citrus flavonoids, troxerutin, and ascorbic acid in food supplements and pharmaceuticals by capillary zone electrophoresis, Food Anal. Methods 6–6 (2013) 1561–1567. [8] L. Tonghua, L. Laichun, Quantitative determination of troxerutin injection by HPLC, China Pharmacy 6, 2004-025. [9] L. Zheng, W. Xiaojiang, Z. Yu, Assay of troxerutin in plasma by RP-HPLC, China Pharmacy 2, 2009-020. 213 [10] Y. Gong Jun, L. Ping, Q. Xi Long, X. Min Juan, Q. Qi Shu, W. Cheng Yin, H. Xiao Ya, W. Zhi Yue, The simultaneous separation and determination of six flavonoids and troxerutin in rat urine and chicken plasma by reversed-phase highperformance liquid chromatography with ultraviolet–visible detection, J. Chromatogr. B 856–1 (2007) 222–228. [11] X. Yang, F. Wang, S. Hu, The electrochemical oxidation of troxerutin and its sensitive determination in pharmaceutical dosage forms at PVP modified carbon paste electrode, Colloids Surf. B: Biointerfaces 52–1 (2006) 8–13. [12] H. Guo, L. Wang, K. Bi, Y. Sun, Determination of troxerutin in troxerutin tablets by monolithic capillary electrochromatography, J. Liq. Chromatogr. Related Technol. 28–5 (2005) 647–658. [13] W. Hong-fu, L. Fang, Determination of carbazochrome sodium sulfonate for injection by UV spectrophotometry, Anhui Medical and Pharmaceutical Journal 1, 2008-010. [14] J. Biao, W. Dong-kai, G. Fei, X. Song-lin, L. Lai, Determination of the content of carbazochrome sodium sulfonate injection and its related substances by HPLC, Journal of Shenyang Pharmaceutical University 3, 2006-007. [15] Z. Ronggui, L. Wei, M. Hongyan, Determination of carbazochrome sodium sulfonate in carbazochrome sodium sulfonate injection by HPLC, China Pharmacy 15 (2005) 24. [16] N. Feng, Y. He, Z. Wang, Chemiluminescence determination of carbazochrome salicylate in sodium periodate-carbazochrome salicylate system, Chin. J. Anal. Lab. 19–6 (2000) 56–58. [17] J. Berzas Nevado, J. Rodriguez Flores, M. Villaseñor Llerena, Simultaneous determination of quinoline yellow and sunset yellow by derivative spectrophotometry and ratio spectra derivative, Anal. Lett. 27–5 (1994) 1009–1029. [18] M.G. El-Bardicy, H.M. Lotfy, M.A. El-Sayed, M.F. El-Tarras, Smart stabilityindicating spectrophotometric methods for determination of binary mixtures without prior separation, J. AOAC Int. 91–2 (2008) 299–310. [19] A. Afkhami, M. Bahram, Mean centering of ratio spectra as a new spectrophotometric method for the analysis of binary and ternary mixtures, Talanta 66–3 (2005) 712–720.