509 Simultaneous Determination of Benzodiazepines by Ultraviolet - Visible Spectrophotornetry in Micellar Media M. de la Guardia, M. V. Galdu, J. Monzo and A. Salvador Departamenio de Quimica Analiiica, University of Valencia, C/Dr. Moliner 50, Burjassot 46700, Valencia, Spain A method for the simultaneous determination of benzodiazepines in binary mixtures is proposed, based on the acid hydrolysis of benzodiazepines to benzophenones and the spectrophotometric determination of the latter in the presence of Nemol K 1030, a non-ionic surfactant condensate of ethylene oxide with nonylphenol. The experimental conditions for the hydrolysis of several benzodiazepines in sealed Pyrex tubes were determined. The addition of Nemol K 1030 to acidic solutions of benzophenones modified the positions of the absorption bands and made possible the simultaneous analysis of binary mixtures of benzodiazepines.Keywords : Benzodiazepines; benzophenones; hydrolysis; surfactants; UV spectrophotometry Benzodiazepines are psychotherapeutic drugs widely used in the treatment of different nervous diseases such as anxiety, insomnia and epileptic convulsions, owing to the wide variety of the properties of the benzodiazepinic structure. 1-3 The therapeutic interest in these compounds justifies research to establish analytical methods for the determination of benzo- diazepines in pharmaceutical preparations and biological samples. Methods proposed for the determination of benzodiaze- pines include liquid chromatography,"5 polarography,h,7 atomic absorption spectrometry8 and molecular spectroscopic techniques such as f-luorimetryy.1(' and infrared" and UV - visible spectrophotometry.For the spectrophotometric determination of this type of compound either the benzodiazepine itself or the benzophe- none obtained by hydrolysis can be employed.12-13 Other methods involve the use of colour reagents that react with these compounds to form species that absorb in the visible region. 14.15 The acid hydrolysis of benzodiazepines to benzophenones has been reportedl.16 and used for the chromatographic determination of these compounds in biological matrices. The hydrolysis is carried out with hydrochloric acid at elevated temperature. One of the methods proposed for obtaining 5-chloro-2-(methylamino)benzophenone (CMAB)l7 involves the treatment of diazepam in stoppered tubes with 6~ hydrochloric for 1 h at 100°C in a boiling water-bath, but in the paper cited the experimental conditions of the hydrolysis reaction were not discussed.In this work we have extended the acid hydrolysis in closed tubes to a wide variety of benzophenones and studied the effect of various experimental conditions. Ultraviolet spectrophotometry is not a selective method because many benzodiazepines have very similar spectra. The use of organised media, such as micellar media, can improve the sensitivity and selectivity of spectrophotometric methods by providing hyperchromic and bathochromic shifts in the spectra of metallic complexes and organic molecules. 18-20 Previous work has demonstrated that the addition of anionic surfactants to solutions of some benzodiazepines in dilute sulphuric acid improves their spectrofluorimetric deter- mination.However, the increase in the fluorescence of benzodiazepines in micellar media is due to the increase in the fluorescence quantum yield without a simultaneous increase in the molecular absorption bands.21-22 In this paper, a method is proposed for the determination of benzodiazepines in binary mixtures based on their hydrolysis to benzophenones and the spectrophotometric determination of the latter in the presence of a non-ionic surfactant. Experimental Apparatus and Reagents A Shimadzu UVi240 UV - visible spectrophotometer equipped with 1-cm quartz cells and a PR-1 graphic printer was used for the absorbance measurements. Powdered samples of diazepam, oxazepam, potassium chlorazepate, temazepam, prazepam, nitrazepam, clonaze- pam and lorazepam were obtained from the Clinical Hospital, Madaus Cerafarm and the Bromatology Department of Valencia University.The following surfactants were used: sodium dodecyi sulphate (SDS) (Fluka), cetyltrimethylammonium bromide (CTAB) (Merck), Descoxid 728, a fatty acid condensate with ethylene oxide (Tensia Surfac), Nemol K 539, 1030 and 1032 condensates of ethylene oxide with nonylphenol and Genapol PF 80 condensate of ethylene oxide and propylene oxide (Hoestch Iberica) and Triton X-100 condensate of tert-octyl- phenol with ethylene oxide (Probus). Hydrolysis of Benzodiazepines Solutions containing 125 pg of diazepam or oxazepam in dilute hydrochloric acid (1 + 1) were used to study the conditions for the hydrolysis of benzodiazepines to benzophenones.Experimental parameters such as temperature, time and acid volume were modified using sealed Pyrex tubes to carry out this reaction in all instances. The recommended procedure consists of placing 1 4 ml of the solution of a benzodiazepine or mixtures of benzodiaze- pines in 6 M hydrochloric acid in a 15-ml Pyrex tube. The tube is hermetically sealed and allowed to stand in an oven at 100-120 "C for 1-1.5 h, allowed to cool and then the contents are diluted to an appropriate volume. Diazepam and oxazepam have strong absorption bands at 240 and 235 nm, respectively, and two other common bands at 285 and 360 nm. The related benzophenones CMAB and 2-amino-5-chlorobenzophenone (ACB) both have a maxi- mum absorbance at 260 nm (see Fig.2 ) . The hydrolysis reaction was monitored by spectrophotometric measurements in the UV region. Determination in the Presence of Surfactants A series of cationic, anionic and non-ionic surfactants were added to the CMAB solutions obtained after hydrolysis of different amounts of diazepam. Calibration graphs in the ultraviolet range were obtained in both the presence and absence of surfactants to determine the effect on the intensityAKALYST, APRIL 1989, VOL. 114 A A 250 3CO 350 400 C 250 300 350 400 hln m Ultraviolet - visible spectra of diazepam, oxazepam and their related benzophenones. Spectra were obtained for 5 p.p.m. solutions of (A) diaze am and (C) oxazepam and the correspondin benzo- phenones, 6) 5-chloro-2-(methylamino)benzophenone (SCMAB) and (D) 2-amino-5-chlorobenzophenone (ACB) and shift of the absorption bands.A 1% surfactant concentra- tion was used in all instances except for solutions of CTAB, for which a 0.05% solution was used owing to its low solubility. Simultaneous Analysis of Mixtures of Benzodiazepines Binary mixtures of benzodiazepines were hydrdysed using the experimental conditions determined previously and the UV spectra recorded in the presence of 1% Nemol K 1030. The absorbance values at the maximum absorbance wavelengths of each benzodiazepine make it possible to determine the concentration of each compound in the mixture using the experimental data obtained for solutions containing only one of the compounds concerned and solving simultaneous equa- tions.Results and Discussion Hydrolysis of Benzodiazepines When samples of diazepam or oxazepam are heated in an oven for 1 h with 1 ml of 6 M hydrochloric acid in pressurised Pyrex vessels, the absorbance of the bands corresponding to benzophenones (260 nm) increases and the absorbance of those of the benzodiazepines (240 and 235 nm) decreases as the temperature is increased. At temptratures higher than 100 "C a constant absorbance was obtained in both instances (see Fig. 2). Using this temperature, the absorbance of the benzophenones obtained by hydrolysis of diazepam and oxazepam increases as the digestion time increases. For a time shorter than 1 h the hydrolysis is not complete, but for longer times the absorbance values remain constant (Fig.3). A volume of 6~ hydrochloric acid greater than 0.75 mi is required for the hydrolysis of diazepam and oxazepam under the previously determined conditions, as can be seen in Fig. 4. Spectroscopic Characteristics of Benzophenones Using the experimental conditions determined previously, the hydrolysis of a series of benzodiazepines was carried out and the following benzophenones were obtained: CMAB from diazepam and temazepam, ACB from oxazepam and potas- sium chlorazepate, 2-amino-S,2'-dichlorobenzophenone (ADCB) from lorazepam, S-chloro-2-(cyclopropylmethyl- amino)benzophenone (CCMB) from prazepam, 2-amino-5- nitrobenzophenone (ANB) from nitrazepam and 2-amino-2'- chloro-5-nitrobenzophenone (ACNB) from clonazepam. The corresponding absorbance maxima and molar absorptivities are given in Table 1.These data indicate the existence of strong overlapping of the absorption bands corresponding to 0.4 - a c 4 0.3 - n Q: $ 0.2 - 0.1 ' I I 20 60 100 140 50 80 110 140 TemperaturePC Fig. 2. Effect of temperature on the acid hydrolysis of diazepam and oxazepam. (a) (U) Absorbance at 240 nm and (@) absorbance at 260 nm of a 5 p.p.m. solution of diazepam after treatment for 1 h with 1 ml of 6~ hydrochloric acid, ( b ) (m) Absorbance at 235 nm and (@) absorbance at 260 nm of a 5 p.p.m. solution of oxazepam treated for 1 h with 1 ml of 6~ hydrochloric acid at different temperatures 0.4 a 0.3 2 2 0.2 I , I I I 0.1 ' 15 ' 45 75 105 15 45 75 105 Timeimi n Fig. 3. Effect of time o n the hydrolysis of diazepam and oxazepam. (a) (M) Absorbance at 240 nm and (@) absorbance at 260 nm of a 5 p.p.m. solution of diazepam after treatment at 100 "C with 1 ml of 6 M hydrochloric acid.( b ) (U) Absorbance at 235 nm and (0) absorbance at 260 nm of a 5 p.p.m. solution of oxazepam after treatment at 100°C with 1 ml of 6~ hydrochloric acid I I I 0 0.5 1 0 1 2 0.1 I VHcliml Fig. 4. Effect of the volume of hydrochloric acid used on the hydro1 sis of diazepam and oxazepam. (a) (H) hbsorbancc at 240 nm and (d) absorbance at 260 nm of a 5 p,p.m. solution of diazepam treated at 100 "C for 1 h with different amounts of hydrochloric acid. (b) (M) Absorbance at 235 nm and (@) absorbance at 260 nm of a 5 p.p.m. solution of oxazepam treated at 100 "C for 1 h with different volumes of 6 M hydrochloric acid the different benzodiazepines, which makes their differentia- tion difficult. Determination of Beozophenone in Micellar Media Calibration graphs for C,MAB in both the absence and presence of different surfactants were obtained.Table 2 gives the equations corresponding to the systems studied. It can be seen that the absorbance maximum at 260 nm is little affected by the addition of anionic, cationic or non-ionic surfactants. However, when a caticnic surfactant such as CTAB or a non-ionic ethylene oxide condensate was added to CMAB solution, a new absorbance band appeared at 415 nm, which makes possible a more selective determination of this benzo- phenone. The addition of Nemol K 1030 to the different benzophe- nones provides a bathochromic shift of the UV bands except for ANB and ANCB, in all instances good linearity of the Cali bration lines corresponding to each of the benzopheno- nones being obtained (Table 3).ANALYST, APRIL 1989, VOL.114 511 Table 1. Spectroscopic characteristics of benzophenones Benzophenone hrn.3, h m CMAB . . . . . . 260 ACB . . . . . . . . 25 8 ADCB . . . . . . . . 2.50 CCMB . . . . . . . . 260 ANB . . . . . . . . 240 365 ACNB . . . . . . . . 250 355 &/moll-1 cm I 11 100 k 300 10300 k 100 10 100 k 100 10200 k 100 13 800 f 100 14800 k 100 11 600 t 100 10200 k 100 Table 2. Effect of addition of surfactants on the spectrophotometric determination of diazepam as CMAB Concen- Regression tration, Calibration coefficient Surfactan t Yo graph* ( r ) None . . . . . . - = 11.057C + 0.02 0.99995 SDS .. . . . . 1 A,,, = 10.565C + 0.01 0.99995 A 4 1 5 = 1.819C - 0.001 0.99995 CTAB . . . . . . 0.05 A,,, = 11 .S48C - 0.05 0.999995 Nemol K 1030 . . 1 AdI5 = 3.931C + 0.006 0.999995 Genapol PF80 . . 1 A260 = 10.811C + 0.01 0.999995 * A = absorbance; C = concentration in mmol 1 - 1 . Table 3. Influence of Nemol K 1030 micelles on the spectrophotometric determination of benzophenoncs Benzo- Calibration graph Calibration graph Ah/ phenone in acidic media" in micellar media* nm CMAB . . A,,,) = 11.057C + 0.008 (Y = 0.99995) ( r = 0.9995) ACB . . A 2 5 8 = 10.161C + 0.020 ADCB . . AZsn = 10.13SC + 0.020 ( r = 0.99995) (Y = 0.99995) CCMB . . A260 = 10.093C + 0.006 ANB . . A T 6 5 = 14.774C-0.004 (Y = 0.99995) ( r = 0.99995) ACNB . . A 3 5 5 = 10.238C-t 0.003 A420 = 4.361C + 0.004 A190 = 2.540C + 0.009 ( r = 0.995) ( r = 0.9995) ( r = 0.99995) A,,,) = 14.S32C + 0.01 1 + 160 +I32 ( r = 0.9995) A395 = 4.179C - 0.003 + 145 A410 = 4.658C - 0.003 +I50 -5 = 12.17SC + 0.011 0 (Y = 0.999995) ( r = 0.999995) * A = absorbance; C = concentration in mmol 1 - 1 .The new bands obtained in acidic micellar media are the same as those obtained in absolute ethanol, which indicates that the micellar microenvironment has similar characteristics to alcohols. In the micellar media the absorbance bands of some of the benzophenones derived from different benzo- diazepines differ significantly, in contrast to the benzodiaze- pine and benzophenone spectra obtained in acidic media, hence allowing the selective determination of these com- pounds in binary mixtures and the determination of diazepam and its metabolites.Analysis of Mixtures of Benzodiazepines It can be see in Fig. 1 that diazepam and oxazepani have very 4 .? 3 400 450 500 J 5 0 A/n m Fig. 5 . Absorption spectra of CMAB and ACB in acidic and micellar media. (a) Absorption spectra of (1) 5 p.p.m. of CMAB; (2) 5 p.p.m. of ACB; and (3) a mixture of 5 p.p.m. of both CMAB and ACB. ( b ) Absorption spectra in the presence of 1% Nemol K1030 for (4) 10 p.p.m. of CMAB; ( 5 ) 10 p.p.m. of ACB; and (6) a mixture of 5 p.p.m. of both CMAB and ACB Table 4. Analysis of binary mixtures of benzodiazepines. Concentrations in p.p.m. Compound Present Oxazepam . . . . . . . . 2.62 5.24 2.62 1.31 5.24 1.37 2.74 5.48 Diazepam . . . . . . . . 1.23 2.46 1.23 4.92 1.23 2.46 1.23 4.92 Diazepam .. . . 1.23 2.46 1.23 4.92 1.25 2.50 1.25 5.00 Found 2.72 5.66 2.88 1.10 5.97 1.36 2.99 6.02 1.32 2.62 1.31 5.20 1.30 2.51 1.10 5.10 1.28 2.71 1.18 5.22 1.34 2.73 1.40 5.33 Relative difference, Yo Compound -3.8 Diazepam . . . . -8.02 -9.9 + 16.03 - 13.9 +0.7 -9.1 -9.8 -7.3 Chlorazepate . . -6.5 -6.5 -5.7 -5.7 -2.03 + 10.6 -3.7 -4.1 Lorazepam . . - 10.2 +4.1 -6.1 -7.2 -9.2 - 12.0 -6.6 Present . . . 2.70 1.35 2.62 5.24 1.31 2.82 2.82 1.41 . . , 2.54 2.54 5.08 1.27 2.46 2.46 4.92 1.23 . . 2.60 2.60 5.20 1.30 2.60 2.60 5.20 1.30 Found 2.92 1.35 2.62 5.29 1.41 2.96 2.97 1.44 2.70 2.54 5.15 1.14 2.60 2.66 5.41 1.32 2.91 2.44 5.30 1.30 2.59 2.39 5.02 1.20 Relative difference, O/* -8.1 0 0 -0.95 -7.6 -5.0 -5.3 -2.1 -6.3 0 -1.4 + 10.2 -5.7 -8.1 - 10.0 -7.3 -11.9 +6.2 -1.9 0 +0.4 +8.1 +3.5 +7.7512 ANALYST, APRIL 1989, VOL.114 similar absorption spectra in the UV - visible region. The spectra of CMAB and ACB are also very similar and as a consequence it is not easy to determine the individual compounds in a mixture of the two [Fig. 5(a)]. However, when Nemol K 1030 is added to the benzophenone solutions the absorption spectra of the two benzophenones differ substan- tially and the binary mixture could be resolved by solving a conventional system of simultaneous equations [Fig. 5(b)]. The same situation occurs for mixtures of diazepam and lorazepam and diazepam and potassium chlorazepate (see Table 3). Hence the simultaneous determination of these benzodiazepines in binary mixtures was carried out by measuring the absorbance at the maximum absorption wavelength for each compound in micellar media after hydrolysis to the corresponding benzophenone.Table 4 summarises the results obtained and the relative differences between the concentrations found and added. References 1. 2. 3. 4. 5. Schutz, H., “Benzodiazepines,” Springer, New York, 1982. Daudon, M., Pharm. Biol., 1977, 11, 389. Civera, J., Master Thesis, University of Valenica, 1982. Mehta, A. C., Tafanta, 1984, 31, 1. de Silva, J. A. F., and Puglisi, C. V., in Garrett, E. R., and Hirtz, J. L., Editors, “Drug Fate and Metabolism,” Volume 4, Marcel Dekker, New York, 1963, p. 245. 6. Barrett, J., Smyth, W. F., and Hart, J. P., J . Pharm. Pharmacol., 1974, 26, 9. 7. Oelschager, H., Biolectrochem. Bioenerg., 1983, 10, 25. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Gonzalez-Pdrez, C., Gonzilez-Martin, M. J., Herniindez- Mendez, J., and Recio-Robuster, R., Qiiim. Anal., 1986, 5 , 420. Gifford, L. A., Miller, J. N., Bridges, J. W., and Burns, D. T., Talanta, 1977, 24, 273. Rodriguez Procopio, J., Hernandez Hernandez, P., and Hernandez Hernandez, L., Analyst, 1987, 112, 79. Taylor, N. F., Jr., and Randall, J., J. Assoc. Off. Anal. Chem., 1979, 62, 799. Giusiani, M., Poggi, G., and Martinelli, G., Riv. Ztal. Zg., 1981, 41, 258. Seitzinger, R. W. T., Pharma. Weekbl., 1975, 110, 1073. Stevens, H. M., J . Forensic Sci. SOC., 1978, 18, 69. Fernandes-Magalhaes, J., and Gisela Pirds, M., Rev. Brasil. Farm., 1970, 195. Lafargue, P., Meunier, J., and Lemontey, Y., J . Chrumutogr., 1971,62,423. de Silva, J. A. F., Schartz, M. A., Stefanovic, V., Japlan, J., and d’Arconte, L., Anal. Chem., 1964, 36, 2099. Hinze, W. L., in Mittal, K. L., Editor, “Solution Chemistry of Surfactants,” Volume 1, Plenum Press, New York, 1979, p. 79. Pelizzetti, E., and Pramauro, E., Anal. Chim. Acta, 1985,169, 1. Cline Love, L. J . , Habarta, J. G., and Dorsey, J. G., Anal. Chem., 1984, 56, 1133A. de la Guardia, M., and Rodilla, F., J. Mol. Struct., 1986, 143, 493. Rodilla, F., Master Thesis, University of Valencia, 1986. Paper 8/02 781 A Received July 11 th, 1988 Accepted November 2nd, 1988