Automatic extraction-spectrophotometric method for the determination of ambroxol in pharmaceutical preparations

Talanta ELSEVIER

Automatic extraction-spectrophotometric method for the determination of ambroxol in pharmaceutical preparations Tom&

P&rez-Ruiz*, Carmen Martinez-Lozano,

Antonio

Sanz, M. Teresa San Miguel

Received I3 September1995: revised 72 No\,ember 1995:accepted I December 1995

Abstract

The spectrophotometric determination of trace amounts of ambroxol was carried out by liquid-liquid extraction using Bromothymol Blue with a flow-injection system. The determination of ambroxol in the range 8 x 10 ‘-4 x 10 ’ M was possible with a sampling frequency of 40 samples h I, The method was satisfactorily applied to the determination of ambroxol in pharmaceutical preparations. Kr~wortls:

Ambroxol: Bromothymol Blue; Flow-injection; Extraction

I. Introduction Ambroxol. 4-[(2-amino-3.5dibromophenyl-methyl) amino] cyclohexanol hydrochloride. is a highly substituted aniline derivative metabolite of bromohexine which also acts as a bronchial mucolytic agent. It is administered as the hydrochloride in daily doses of 30P 120 mg and is available commercially as syrups. tablets and granules. Similar doses have been given by inhalation, injection or rectally, producing good results in the treatment of chronic bronchitis and alveolar proteinosis. Several different methods have been used for the determination of ambroxol, including ultravi* Corresponding author

olet spectroscopy [l]. thin layer chromatography [2], gas-liquid chromatography [ 1,3,4] in conjunction with electron capture and high performance liquid chromatography using UV and amperometric detection [I.591. No automatic analytical methods for ambroxol that permits its determination (e.g. continuous flow analysis or flow injection) have been reported. Flow-injection (FI) methodology in association with extraction in organic solvents currently provides a means of automating and speeding up the handling of reagents in routine analysis with good selectivity and sensitivity. The purpose of this work was to investigate systematically the formation and extraction behaviour of ion pairs of ambroxol with acid dyes in order to develop useful automatic photometric methods. The results showed that Bromothymol

0039-9140~96 $15.00C 1996ElsevierScienceB.V. 411rightareserved PII soo39-9 l-lO(Y6lO 1850-4

Blue and 1,2-dichloroethane were the most eflective dye and extractant respectively, for use in unsegmented flow configurations using a continuous extraction system. This system overcame the complexity of the manual extraction methods and avoided the troubles and hazards involved in handling toxic organic solvents. Flow injection minimizes the above shortcomings as the organic solvents are kept in closed vessels. The proposed automatic method has been applied to the determination of ambroxol in pharmaceutical preparations.

2. Experimental 2.1. Rrugents Ambroxol was obtained from Sigma (St. Louis. MO) and used as received. A standard 1.0 x 10 ’ M solution was prepared by dissolving the drug in distilled water: this solution remained stable for 2 weeks if kept refrigerated. Working solutions of lower concentrations were freshly prepared by appropriate dilution of the standard solution. A 1.O x 10 ~ ’ M Bromothymol Blue (4-4’-3H2,l -benzoxathiol-3-ylidene)bis[2-bromo-3-methyl6-( 1-methylethyl)phenol]-S,S-dioxide, stock solution was prepared by dissolving the required amount of dye (Sigma) in water. Solutions of lower concentration were prepared by dilution of the stock solution with distilled water. 2.2. Appuratus A Perkin-Elmer (Norwalk, CT) 550 SE spectrophotometer was used for recording spectra and a Pye-Unicam (Cambridge, UK) 8625 spectrophotometer was used as the detector in the flow system. A Gilson (Villiers le Ball, France) Minipuls HP4 peristaltic pump fitted with Tygon and Acidflex pump tubes and an Omnifit (Cambridge, UK) injection valve were also used.

The configuration fold used is depicted

of the flow-injection maniin Fig. 1 with the optimum

conditions as stated. Acetate buffer and Bromothymol Blue solutions were pumped through Tygon tubes and 1.2-dichloroethane was pumped through the Acidflex tube. The sample (210 ill) was introduced into the buffer stream by means of an Omnifit rotary valve to which a volume control loop was attached. All connecting tubing was made of poly(tetrafluoroethylene) (PTFE). A Tsegmenter, in which the aqueous phase flows straight and the organic phase at right-angles, was used for mixing both phases. The extraction coil was 150 cm long. The phase separator was constructed from solid PTFE which had an inlet and two outlets (bore 0.5 mm i.d.). The three-threaded hole accepted the standard polypropylene end pieces. During operation the two blocks were pressed together with the aid of two stainless-steel pins. A porous PTFE membrane with 1.0 pm pore size (Fluoropore, Millipore Iberica, Madrid, Spain), permeable to chloroform but impermeable to the aqueous solution, was sandwiched between the two blocks. A grid placed between the membrane and the inside non-grooved surface of the block prevented the membrane from collapsing into the recipient chamber, the volume of which was only 20 111. The absorbance of the organic phase was measured at 420 nm with a spectrophotometer equipped with a Hellma (Jamaica, NY, USA) 178.012 QS flow cell (18 ~1 inner volume and 10 mm light pathlength) and was recorded with a Linseis (Selb, Germany) 6215 recorder. The Acidflex pump tubing was found to be of varying quality; the lifetime varied from a few hours to 2 weeks. Since the flow rates of the PP

Fig. I. Manifold for determination of ambroxol: PP, peristaltic pump; R,, sample: R,. acetate butTer (pH 4): R,, Bromothymol Blue; R,. l,2-dichloroethane; RC. reaction coil (30 cm x 0.5 mm i.d. ): S. segmenter: PS. phase separator; EC, extraction coil (150 cm x 0.5 mm i.d.): W, waste: D. detector.

T. Pt;rr:-Ruin

et al. : Tulun~u

Table I Elect of the extracting solvent on absorbance ([Ambroxol] 2 x IO-’ M; [Bromothymol Blue] = I x lo-’ M. pH 4) Solvent

A 1”” pt” -A bl.LnL

Relative signal (‘%)

I ,2-Dichloroethane Chloroform Toluene Ethyl acetate Carbon tetrachloride Isobutyl methyl ketone

0.669 0.427 0.151 0.074 0.129 0.124

100 64 23 II I9 18

=

organic streams change because of mechanical deterioration of the Acidflex tubings, three standard solutions were introduced every day. The Fluoropore membrane had a mean lifetime of 20 h; however, the peak heights remained constant when the membrane was changed.

3. Results and discussion Ambroxol (AM) can be transferred from the aqueous phase into the organic phase as an ion pair formed with the anionic form of the acid dyes, The following steps take place: Am,&, + D,,,

ti Am + D,,, F! Am + D (OV2)

4.1 (1996)

10%

1034

1031

3.1. Charactrristics of’ the Ambroxol- Brormthyttzol Blue ion puit Bromothymol Blue and the ion pair Am + D have identical spectra and so they must be separated if the ion pair is to be quantified. The effect of pH on the formation and extraction of the ion pair was studied using universal buffer BrittonRobinson solutions over the range 1.5-6.0. The absorbance of the organic extract was maximal and constant in the pH range 334.2 (Fig. 2). The composition of the ion pair was established by Job’s method of continuous variations [lo] and by the molar ratio method [1 I] using both variable dye concentration and variable ambroxol concentration. The results obtained with these methods showed that the composition of the associate was equimolar (1: 1). The extraction constant for the equilibrium (1) was log K,, = 5.6 k 0.4. Shaking times ranging from 0.5 to 5 min did not produce any change in the absorbance, suggesting that equilibrium between the two phases in the extraction of the ion pair can be attained rapidly. Reproducible absorbance readings were always obtained after a single extraction. The overall extraction efficiency was 96.8%

(1)

where Am+ and D ~ represent the protonated ambroxol and the anion of the dye. The dyes studied for ambroxol ion-pair formation were Chromazurol S, Orange IV, Methylthymol Blue, Methyl Orange and Bromothymol Blue. Of the dyes tested. Bromothymol Blue showed the greatest ion-pair extraction efficiency with the smallest reagent blank extraction. The effect of the extraction solvent used, was also examined. The polarity of the solvent affects both the extraction efficiency and the absorbance. The results using Bromothymol Blue are shown in Table 1, in which the response using 1,2dichloroethane was normalized as 100. In this study, 1,2-dichloroethane was preferred to chloroform because of its lower volatility.

Fig. 2. Influence broxol] = 5 x IO

of pH on the extraction of ion pair. [Am‘: [Bromothymol Blue] = 5 x IO ’ M.

7“alarzta

43 (1996)

1029- 1034

because the ion pair is formed rapidly. The influence of the extraction coil length was also examined. The peak height increased as the extraction coil increased in length up to 80 cm, remained constant between 80 and 200 cm and decreased at greater lengths. An extraction coil length of 150 cm (0.5 mm i.d.) was selected. The volume of sample injected was varied from 35 to 250 ~1 by changing the length of the sample loop in the injection valve. The peak height increased with increasing sample size up to 200 ~1, above which it remained virtually constant. The volume to be injected was selected as 210 ,~tl. 1 1

3

I J

[Bromthymol

Blur],

Fig. 3. Etkxt of pH (0) and Bromothymol (‘2) on the peak height.

3.2. Flo\t,-injection

7

7

M x 10.

Blue concentration

a’eterminution umhroxol

The flow manifold (Fig. 1) for the automation of the proposed method was arranged so as to consider the essential features of the ambroxolL Bromothymol Blue ion pair.

Injhence of nimi~bld paranietrrs The optimization of the manifold parameters with respect to sensitivity, peak resolution, phase separation efficiency and rapidity of the analysis was carried out using the results obtained from the batch studies. The carrier was an acetate buffer pH 4 (0.5 M) and the reagent stream was Bromothymol Blue solution (6 x 10 ’ M). The sample solution used was 6 x 10 ’ M. The flow rates of the aqueous and organic streams were varied in order to obtain the maximum concentration coefficient without significantly decreasing the sample throughput. The optimization of the flow rate resulted in the adoption of flow rates of 1.6 (0.8 for each channel) and 2.0 ml min- ’ for the aqueous and organic streams respectively. The tube length between the valve and segmenter (ion-pair reaction coil) was varied from 20 to 60 cm (0.5 mm Cd.). A reaction coil of 30 cm was sufficient to obtain the maximum absorbance 3.21.

3.22. EJkct of‘ the reagent concentration With the concentration of Bromothymol Blue solution fixed at 6 x 10 -~4 M the pH of the buffer solution (carrier) was varied between 2.0 and 5.0. The peak height was maximal and constant from pH 3.2-4.5. and decreased outside this range (Fig. 3). Therefore, a 0.5 M acetate buffer of pH 4.0 was used as carrier. With pH fixed at 4.0 the concentration of Bromothymol Blue was varied between 5 x lo--’ and 8 x 10-j M. The peak height increased with increasing concentrations of the dye solution stream up to 3 x lop4 M, but levelled off at higher concentrations. The concentration adopted in the procedure was 6 x 10 - 4. 3.23. Ambroxol determination The effect of the concentration of ambroxol on the absorbance was studied by measuring the peak height when 210 ~1 portions of ambroxol Table 2 Tolerance broxol” Species

of different

species

in the determination

added

Fructose. tartrate,

Maximum mole ratio

lactose. sucrose, glycerine. zinc(lI), urea, sorbitol

lOOh

Saccharine. oxalic acid. citric acid, benzoic acid, propylene glycol

50

Polyethylene

glycol.

0.5

d [Ambroxol] h Maximum

= 5 x 10~ ‘. ratio tested.

bromhexine

of am-

tolerable

1033 Table 3 Determination

of ambroxol

Preparation”

in pharmaceutical

preparations

Supplier

Found Proposed

Motosol (packet) granulated Mucosan (syrup) Naxpa (syrup) Mucibron (solution)

Europharma Fher Novag Hosbon

method

’

61 ? 1 (mg ml I) 3.O*O.l (mg ml-‘) 3.0*0.1 (mg ml-‘) 3.0 * 0.1 (mg ml-‘)

Reference

method

61 i

1 (m&9 2.90 +O.l (mg ml-‘) 3.O+O.l (mg ml-‘) 3.1 kO.1 (mg ml-‘)

J Composition of samples: Motosol granulated: ambroxol chlorhydrate 60 mg, saccharin sodium 12 mg. sorbitol, excipients; Mucosan: ambroxol chlorhydrate 15 mg. excipient in 5 ml: Naxpa: ambroxol chlorhydrate 300 mg, excipient in 100 ml; Mucibron: ambroxol chlorhydrate 300 mg. excipient in 100 ml b Mean i standard deviation for four samples. c UV absorbance [I].

hydrochloride solutions of different concentrations were injected. The calibration graph was found to be linear between 8.0 x 10Ph and 4.0 x 10 P4 M. and the regression equation obtained was: A = (0.013 k 0.006) + (2891 k 45)C; (r = 0.9987)

(2)

where C is the molar concentration of ambroxol, A is the absorbance and r is the correlation coefficient. The relative standard deviations of 10 injections of each solution containing 1.0 x 10 ’ and 8.0 x 10 - 5 M ambroxol were 0.88 and 0.32% respectively. The detection limit, calculated as the value corresponding to three times the standard deviation of the blank, was 5 x 10 ~ 7 M of ambroxol. The sampling rate was 40 samples h ~ ’ The reproducibility of the method was studied by analysing, on five different days, 10 identical solutions of ambroxol (8.0 x lo-’ M). Every day three injections of each solution were made; the relative standard deviation of the peak height was 1.34%. The selectivity of the method was studied by analysing synthetic sample solutions containing 5 x 1’OP5 M of ambroxol and various excess amounts of foreign substances. The tolerance ratio of each foreign compound was taken as the largest amount yielding an error less than k 4%

in the analytical signal of ambroxol. Table shows the results obtained in this study.

2

3.2.4. Anul~~sis of pharnzaceuticul preparations In order to establish the validity of the proposed method, several pharmaceutical preparations were analysed. Interference from excipients present was not a problem. The data in Table 3 show that the assay results were in good agreement with values for the nominal contents and with those using the manual photometric reference method [l]. The recoveries obtained for ampharmaceutical broxol added to each formulation were in the range 97.2-102.6%.

4. Conclusions Results of experiments with different dyes and extraction solvents showed that Bromothymol Blue and 1,2-dichloroethane respectively were the most effective for use in an unsegmented flow configuration with a continuous extraction system. This system overcame the complexity of normal extraction methods and avoided problems and hazards involved in handling toxic organic solvents. The sensitivity, dynamic range and throughput of the method are good for the determination of ambroxol in pharmaceutical preparations.

1034

T. P&e:-Rui:

et al. / Talanta

43 (1996)

Acknowledgements

The authors are grateful for the financial support given by the Direccibn General de Investigaci6n Cientifica y Tkcnica (Project PB93-1139).

[4] [5] [6] [7]

References [I] G. Indrayanto and R. Handayani, J. Pharm. Biomed. Anal.. 11 (1993) 781. [2] C.E. Uboh, J.A. Rudy. L.R. Soma, M. Fennel], L. May. R. Sams. F.A. Railing. J. Shellenberger and M. Kahler. J. Pharm. Biomed. Anal.. 33 (1991) 9. [3] L. Colombo. F. Marcucci, M.G. Marini. P. Pierfederici

[8] [9]

[IO] [I 1]

1029-1034

and E. Mussini, J. Chromatogr. Biomed. Appl., 95 (1990) 141. L. Schmid, J. Chromatogr. Biomed. Appl., 58 (1987) 65. M. Nieder and H. Jaeder, J. High Resolut. Chromatogr. Commun.. 9 (1986) 561. M. Nobilis, J. Pastera, D. Svoboda. J. Kvetina and K. Macek, J. Chromatogr. Biomed. Appl., 119 (1992) 251. M.H.A. Botterblom. T.J. Janssen, P.J.M. Guelen and T.B. Vree, J. Chromatogr. Biomed. Appl., 65 (1987) 211. V. Brizzi and U. Passeti, J. Pharm. Biomed. Anal.. 8 (1990) 107. F.J. Flores-Murrieta. C. Hoyo-Vadillo, E. Hong and G. Castaneda-Hernandez J. Chromatogr. Biomed. Appl., 82 ( 1989) 464. H. Irving and T.B. Pierce. J. Chem. Sot.. (1959) 2565. J.H. Yoe and A.L. Jones Ind. Eng. Chem. Anal. Ed., 16 (1944) 111.