Comparison of two derivative spectrophotometric methods forthe determination of a-tocopherol in pharmaceutical preparations

Bilal Yılmaz, Murat Öztürk, Yücel (Yasar) Kadıoglu *

Department of Analytical Chemistry, Faculty of Pharmacy, Atatürk University, Erzurum 25240, Turkey

Received 4 December 2003; received in revised form 15 April 2004; accepted 9 May 2004

Available online 28 July 2004

Abstract

Simple, sensitive and reliable derivative spectrophotometric methods were developed and validated for determination of a-tocopherol inpharmaceutical preparations. The solutions of standard and the sample were prepared in absolute ethanol. The quantitative determination ofthe drug was carried out using the first derivative values measured at 284, 304 nm and the second derivative values measured at 288, 296 nm.Calibration graphs constructed at their wavelengths of determination were linear in the concentration range of a-tocopherol using peak to zero10–250 µg ml–1 for first and second derivative spectrophotometric methods. Developed spectrophotometric methods in this study are accurate,sensitive, precise, reproducible, and can be directly and easily applied to Evon dragee form as pharmaceutical preparation. Statistical analysis(Student’s t-test) of the obtained results showed no significant difference between the proposed two methods.© 2004 Elsevier SAS. All rights reserved.

Keywords: a-Tocopherol; Vitamin E; Pharmaceutical dosage form (dragee); Derivative UV spectrophotometry

1. Introduction

Vitamin E is the collective name for the eight majornaturally occuring molecules, four tocopherols (a, b, c, d)and four tocotrienols (a, b, c, d). Tocotrienols differ fromtocopherols in that they have an isoprenoid instead of aphytyl side chain (Fig. 1). The four isoforms of tocopherolsand tocotrienols differ in the number of methyl groups on thechromanol nucleus (a has three, b and c have two, while dhas one) [1–2]. a-Tocopherol is present in human skin, par-ticularly in the epidermis, which is the outermost skin layer,working as an effective photoprotective agent and function-ing as a primary antioxidant in the first line of defense againstharmful reactive oxidant species [3].

The various vitamin E isoforms possess important physi-ological roles beyond their antioxidant activities [4–6]including hypocholesterolemic, antithrombotic, anti-inflam-matory and antiproliferative effects.

In addition, a-tocotrienol was recently shown to act as apotent modulator of src-kinase signal transduction that wasnot sensitive to comparable concentrations of a-tocopherol[7].

Evon dragee is a powerful antioxidant, protects your cellsfrom oxidation, and neutralizes unstable free radicals, whichcan cause damage. This is done by the vitamin E giving upone of its electrons to the electron deficient free radical,making it more stable. While vitamin E performs its antioxi-

* Corresponding author.E-mail address: ykadiogli@yahoo.com (Y. (Yasar) Kadıoglu). Fig. 1. Chemical structure of tocopherols and tocotrienols.

IL FARMACO 59 (2004) 723–727

www.elsevier.com/locate/farmac

© 2004 Elsevier SAS. All rights reserved.doi:10.1016/j.farmac.2004.05.004

dant functions, it also protects the other antioxidants frombeing oxidized. Evon dragee contains vitamin E, lactose,cocoa, sugar, titandioxide, and b-carotene.

Several methods have been reported for the determinationof a-tocopherol. Voltammetric [8] and HPLC methods for thedetermination of a-tocopherol have largely replaced directspectrophotometric and fluorometric procedures [9]. HPLChad been used with fluorescence detection [10–15], with UVdetection [16–19], and most recently with detection byevaporative light scattering [20].

Among the various methods available for the determina-tion of drugs, spectrophotometry continue to be popular,because of their simplicity and specificity.

The proposed methods are cheaper and more simple thanHPLC methods. It might be an alternative to the HPLCtechniques for routine analysis and there are no extractionprocesses to eliminate the excipients, which are time con-suming and tedious.

The aim of this study is to develop two simple, sensitiveand validated, derivative UV spectrophotometric methodsfor the determination of a-tocopherol and to apply thismethod to the commercial pharmaceutical preparations. Theresults obtained by these two methods were statistically com-pared.

2. Experimental

2.1. Apparatus

Thermospectronic double beam UV–Vis spectrophotom-eter (HEkIOSb, Thermospectronic, Cambridge, UK) withthe local control software was used. UV- and derivativespectra of the solutions were recorded in 1-cm quartz cells ata scan speed of 600 nm min–1, a scan range of 270–320 nm,fixed slit width of 2 nm, and derivation interval of (Dk)1.0 nm.

2.2. Reagent and standards

a-Tocopherol was purchased from sigma (St. Louis,MO,USA). It was tested for purity by controlling its UV andinfrared spectra, and the purity was found to be approxi-mately 95%. Evon dragee, containing about 200 mg ofa-tocopherol was purchased from local market (Erzurum,Turkey). Absolute ethanol was purchased from sigma(Seelze, Germany).

2.3. Solutions

For first and second derivative spectrophotometric analy-ses, stock solution of a-tocopherol was prepared at a concen-tration of 300 µg ml–1 in absolute ethanol. Working standardsolutions were daily prepared by diluting the stock solutionat the concentrations of 10, 50, 100, 150, 200, and 250 µgml–1 with absolute ethanol in volumetric flasks (10 ml).Absolute ethanol was used as a blank solution.

2.4. Assay procedure

The mean weight of finely powdered Evon dragee con-taining 200 mg of a-tocopherol was accurately transferredinto 100-ml calibrated flask and 75 ml of absolute ethanolwas added.

The flask was placed in ultrasonic vibration for 50 min atroom temperature and diluted with absolute ethanol to mark.The mixture was filtered through nylon membrane of0.45 µm pore size (47 mm filter membrane, Phenomenex,USA). One milliliter of this solution was transferred to the10-ml calibrated flask and diluted with absolute ethanol tothe mark.

2.5. Data analysis

All statistical calculations were performed with the statis-tical product and service solutions (SPSS) for windows,version 10.0. Correlations were considered statistically sig-nificant if calculated P-values were 0.05 or less.

3. Results and discussion

3.1. Spectrophotometric methods

UV spectrum of a-tocopherol in absolute ethanol gavetwo broad shouldered peaks with maximum wavelengths at291 and 296 nm, respectively (Fig. 2). These maximumwavelengths were broader at low concentrations so thatanalysis could not be performed. Derivative UV spectropho-tometry was preferred for the analysis of a-tocopherol sincethe amplitude of the signal of derivative spectra was greater,the peak shape was well defined, and the separation of theshouldered peaks was better in this method.

The first- and second-order derivative UV spectrum analy-sis of a-tocopherol gave sharper and better-defined peakswhen compared with the zero-order derivative spectrum ofa-tocopherol (original) (Figs. 3,4).

Fig. 2. UV spectrum of a-tocopherol (200 µg ml–1 a-tocopherol in absoluteethanol).

724 B. Yılmaz et al. / IL FARMACO 59 (2004) 723–727

First-order derivative spectrum of a-tocopherol showedtwo maxima at 284 and 294 nm and a minimum 304 nm(Fig. 3). Second-order derivative (2D) curve displayed threemaxima at 281, 293, and 308 nm and two minima at 288 and296 nm against blank as a reference substance (Fig. 4).a-tocopherol was determined by means of first derivative ofUV-absorbance values (A) at 284 and 304 nm and fromvalues from the second derivative of UV-spectra (2D) at288 and 296 nm.

3.2. Method validation

3.2.1. LinearityFor first and second derivative spectrophotometric analy-

sis, linearity ranges were to be 10–250 µg ml–1 in absoluteethanol. For first-order derivative regression equations wereA = 0.0031x + 0.069 (x: concentration, r = 0.995) andA = 0.0024x + 0.029 (x: concentration, r = 0.996) at 284 and304 nm, respectively.

For second-order derivative calibration curves were con-structed by plotting 2D values against concentrations (10–250 µg ml–1) of a-tocopherol standards in absolute ethanol.Regression equations of linear calibration graphs were calcu-lated as 2D = 0.0033x + 0.036 (x: concentration, r = 0.998)and 2D = 0.0031x – 0.040 (x: concentration, r = 0.996) at288 and 296 nm, respectively.

In Table 1, the statistical parameters are given; the regres-sion equations are calculated from the calibration graphs,along with the standard deviations of the slope (Sb) andintercept (Sa) on the ordinate.

3.2.2. SpecificityComparison of the first and second derivative spectrum of

a-tocopherol in Evon dragee solutions show that the wave-length of maximum absorbance did not changed. Accordingto the results obtained, the derivative spectrophotometricmethods are able to access the analyte in presence of excipi-ents and hence, they can be considered specific.

3.2.3. Accuracy and recoveryRecovery experiments were conducted to determine the

accuracy of the proposed methods. Recovery studies wereperformed at three concentration levels (25, 75, and 125 µgml–1). The mean recovery and relative standard deviation(RSD) were presented in Tables 2 and 3. The results obtainedusing the above methods were accurate and precise.

3.2.4. RepeatabilityRepeatability is given as intra- and inter-day precisions

were evaluated by analyzing the three different concentra-tions and three different days of pure a-tocopherol. Theinter-day precision was evaluated by comparing the linearregressions of three standard plots prepared on three different

Fig. 3. First derivative spectrum of a-tocopherol (200 µg ml–1 a-tocopherolin absolute ethanol).

Fig. 4. Second derivative spectrum of a-tocopherol (200 g ml–1 a-tocopherolin absolute ethanol).

Table 1Analytical data for the calibration graphs (n = 6) for the determination of a-tocopherol by the proposed methods

Parameters Derivative spectrophotometryFirst Second

k (nm) 284 304 288 296Range (µg ml–1) 10–250 10–250 10–250 10–250Regression equation A = 0.0031x + 0.069 A = 0.0024x + 0.029 2D = 0.0033x + 0.036 2D = 0.0031x – 0.040Sa 0.0084 0.0057 0.0038 0.0066Sb 0.0002 0.0007 0.0009 0.0012Correlation coefficient (r) 0.995 0.996 0.998 0.996

Sa, standard deviation of intercept of regression line.Sb, standard deviation of slope of regression line.x is the concentration in µg ml–1, A and 2D are the amplitude for first and second derivatives respectively.

725B. Yılmaz et al. / IL FARMACO 59 (2004) 723–727

days. Six replicate determinations at three different concen-trations for pure a-tocopherol were carried out to test theprecision of the proposed methods. The results are given inTables 2 and 3. The precision of the assays is demonstratedby RSD of lower than 1.99%.

3.2.5. RuggednessIn this study, first and second derivative UV spectropho-

tometric determination of a-tocopherol were carried out by adifferent analyst in same instrument with the same standard(Table 4). The results showed no statistical differences be-tween different operators suggesting that the developedmethods were rugged.

3.2.6. StabilityStock solutions of a-tocopherol were prepared at a con-

centration of 300 µg ml–1 in absolute ethanol and kept in the

dark at +4 °C. Stability of a-tocopherol stock solutions weretested during 2 weeks and results showed that a-tocopherolsolutions in absolute ethanol were stable in this period.

4. Conclusion

In the present work, two new derivative spectrophotomet-ric methods have been developed and validated for routinedetermination of a-tocopherol in pharmaceutical prepara-tions. According to the statistical comparison (Student’st-test) of the results there is no significant difference betweenfirst and second derivative spectrophotometric methods(Table 5). The presented methods are concluded as accurate,precise, reproducible, and can be directly and easily appliedto the pharmaceutical formulations of a-tocopherol.

Table 2Intra- and inter-day precision and accuracy of a-tocopherol (n = 6) with first derivative spectrophotometric method

Intra-day Inter-dayAdded(µg ml–1)

Found(Xa)

Precision(% RSD)

Accuracy(%)

Added(µg ml-1)

Found(Xa)

Precision(% RSD)

Accuracyb

(%)First spectrophotometric method (284 nm)

25 25.2 1.32 1.43 25 25.4 1.31 2.0375 75.4 1.43 0.91 75 75.7 1.72 1.60125 125.5 0.43 0.76 125 125.7 0.88 0.68

First spectrophototmetric method (304 nm)25 25.3 1.68 1.82 25 25.5 1.99 2.3975 75.6 0.73 0.79 75 75.8 0.76 1.86125 125.8 0.87 0.86 125 125.7 1.46 0.95

RSD, relative standard deviation.a X, mean values represent six a-tocopherol standard solutions for each concentration.b Accuracy (% relative error): (found – added)/added × 100.

Table 3Intra- and inter-day precision and accuracy of a-tocopherol (n = 6) with second derivative spectrophotometric method

Intra-day Inter-dayAdded(µg ml-1)

Found(Xa)

Precision(% RSD)

Accuracyb

(%)Added(µg ml-1)

Found(Xa)

Precision(% RSD)

Accuracyb

(%)Second spectrophotometric method (288 nm)

25 25.3 1.20 1.20 25 25.5 1.19 2.0075 75.6 1.37 0.80 75 75.8 1.21 1.06125 125.8 0.46 0.64 125 125.9 0.91 0.72

Second spectrophotometric method (296 nm)25 25.4 1.13 1.60 25 25.6 1.99 2.4075 75.7 0.33 0.90 75 75.9 0.78 1.20125 125.9 0.68 0.72 125 126.1 1.34 0.88

RSD, relative standard deviation.a X, mean values represent six a-tocopherol standard solutions for each concentration.b Accuracy (% relative error): (found – added)/added × 100.

Table 4The results of analyses of standard a-tocopherol by a different analyst a

Method Different analystAdded (µg ml-1) Found (µg ml-1) Standard deviation % RSD

First derivative spectrophotometry (284 nm) 175 175.8 2.14 1.22First derivative spectrophotometry (304 nm) 175 175.4 2.35 1.34Second derivative spectrophotometry (288 nm) 175 175.6 3.31 1.89Second derivative spectrophotometry (296 nm) 175 175.9 3.53 2.01

a Found results are mean six separate measurements of first and second derivative spectrophotometric methods.

726 B. Yılmaz et al. / IL FARMACO 59 (2004) 723–727

References

[1] A. Kamal-Eldin, L.A. Appelqvist, The chemistry and antioxidantproperties of tocopherols and tocotrienols, Lipids 31 (1996) 671–701.

[2] G. Pongracz, H. Weiser, D. Matzinger, Tocopherols-antioxidants innature, Fat. Sci. Technol. 97 (1995) 90–104.

[3] J. Fuchs, Potentials and limitations of the natural antioxidants RRR-alpha-tocopherol, L-ascorbic acid and b-carotene in cutaneous photo-protection, Free Radic. Biol. Med. 25 (1998) 848–873.

[4] M.G. Traber, L. Packer, Vitamin E: beyond antioxidant function, Am.J. Clin. Nutr. 62 (Suppl. 6) (1995) 1501–1509.

[5] A. Theriault, J.T. Chao, Q. Wang, A. Gapor, K. Adeli, Tocotrienol: areview of its therapeutic potential, Clin. Biochem. 32 (1999) 309–319.

[6] L. Packer, S.U. Weber, G. Rimbach, Molecular aspects of alpha-tocotrienol antioxidant action and cell signalling, J. Nutr. 131 (2001)369–373.

[7] C.K. Sen, S. Khanna, S. Roy, L. Packer, Molecular basis of vitamin Eaction. Tocotrienol potently inhibits glutamate-induced pp60(c-Src)kinase activation and death of HT4 neuronal cells, J. Biol. Chem. 275(2000) 13049–13055.

[8] V.J. Priya, S.I. Vijaykumar, K.S. Ashwini, Voltammetric behavior ofa-tocopherol and its determination using surfactant+ethanol+waterand surfactant+acetonitrile+water mixed solvent systems, Anal.Chim. Acta 441 (2) (2001) 201–206.

[9] J.F. Gregory, in: O.R. Fennema (Ed.), Química de los Alimentos,Acribia S.A., Rayo, Zaragota, Spain (2000) 633 [Chapter 8].

[10] A. Rodríguez-Bernaldo de Quirós, J. López-Hernandez, J. Simal-Lozano, Determination of carotenoids and liposoluble vitamins in seaurchin (Paracentrotus lividus) by high performance liquid chromatog-raphy, Eur. Food Res. Technol. 212 (2001) 687.

[11] G.W. Chase Jr, A.R. Long, Liquid chromatographic method for analy-sis of all-rac-alpha-tocopheryl acetate and retinyl palmitate in milk-based infant formula using matrix solid-phase dispersion, J. AOACIntern. 81 (1998) 582.

[12] A. Sobczak, B. Skop, B. Kula, Simultaneous determination of serumretinol and a- and c-tocopherol levels in type II diabetic patients usinghigh-performance liquid chromatography with fluorescence detec-tion, Journal of Chromatography B: Biomedical Sciences and Appli-cations. 730 (2) (1999) 265–271.

[13] A.K. Hewavitharana, A.S. van Brakel, M. Harnett, SimultaneousLiquid Chromatographic Determination of Vitamins A, E andb-Carotene in Common Dairy Foods, Int. Dairy J. 6 (1996) 613.

[14] D. Hoehler, A.A. Frohlinch, R.R. Marquardt, H. Stelsousky, Extrac-tion of alpha-tocopherol from serum prior to reversed-phase liquidchromatography, J. Agric. Food Chem. 46 (1998) 973.

[15] S.T. Chou, L.E. Ko, C.H.S.H. Yang, Determination of tissuea-tocopherol in senescence-accelerated mice by high-performanceliquid chromatography with fluorimetric detection, Anal. Chim. Acta419 (2000) 81.

[16] A. Lynch, J.P. Kerry, D.J. Buckley, P.A. Morrissey, C. López-Bote,Use of high pressure liquid chromatography (HPLC) for the determi-nation of cda-tocopherol levels in forage (silage/grass) samples col-lected from different regions in Ireland, Food Chem. 72 (2001) 521.

[17] C.J. Wyatt, S. Pérez, R.O. Méndez, a-and g-tocopherol content ofselected foods in the mexican diet: Effect of cooking losses, J. Agric.Food Chem. 46 (1998) 4657.

[18] S.A. Albalá-Hurtado, S. Novella-Rodríguez, M.T. Veciana-Nogués,A. Mariné-Font, Determination of vitamins A and E in infant milkformulae by high-performance liquid chromatography, J. Chro-matogr. A. 778 (1997) 243.

[19] M.G. Paulo, H.M. Cabral, J.A.G. Morais, A.J. Almeida, An isocraticLC method for the simultaneous determination of vitamins A, C, Eand b-carotene, J. Pharm. Biomed. Anal. 21 (1999) 399.

[20] K. Warner, T.L. Mounts, Analysis of tocopherols and phytosterols invegetable oils by HPLC with evaporative light-scattering detection, J.Am. Oil Chem. Soc. 67 (1990) 827.

Table 5Statistical evaluation of obtained data from two methods in pharmaceutical preparation containing a-tocopherol (200 mg)

Method n Mean Standarddeviation

Standard errormean

CI(min–max)

t-values

First derivative spectrophotometry 12 196.5 4.75 1.12 193.0–199.8 tc = 1.49Second derivative spectrophotometry 12 198.5 4.11 0.97 196.8–200.3 tt = 3.32

n: number of determination, CI: confidence interval, tc: calculated t-value, tt: tabulated t-value, Ho hypothesis: no statistically significant difference existsbetween two methods, tt > tc; Ho hypothesis is accepted (P > 0.05).

727B. Yılmaz et al. / IL FARMACO 59 (2004) 723–727