OCTOBER 1980 The Analyst Vol. 105 No. 1255 Polarographic Behaviour and Analysis of Some Azo Dyes of Biological Significance J. P. Hart Orthopaedic Department, Charing Cross Hospital, London, W.6 and W. Franklin Smyth Chemistry Department, University College, Cork, Republic of Ireland The polarographic behaviour of the azo dyes CI Direct Orange 34, Acid Red 73, Direct Blue 84 and Direct Red 80 over the pH range 1-13 was studied. Mechanisms of reduction are postulated, the optimum pH ranges for the determination of the dyes by differential-pulse polarography are selected and the appropriate linear ranges deduced from calibration graphs. Keywords: Azo dyes; fiolarography Azo compounds are widely used in industry as textile dyes, colouring agents in foods and pharmaceuticals, etc.As a result they can become environmental pollutants through dis- charge of the contents of plating baths from textile works into rivers, especially if not treated by the activated sludge process, as has been observed in certain instances.lV2 They can also enter the body through the intake of certain foods and drugs that contain these azo compounds. Concern has been voiced about the potential carcinogenicity of compounds containing azo group(^).^-^ Therefore, a study of the mechanism of the polarographic reduction of azo compounds, which often parallels the metabolism of these compounds in vivo (e.g., reductive fission of the azo linkage to the parent aromatic amines), is worthy of investigation. Differen- tial-pulse polarography can also be used to determine the parent compounds and their electroactive metabolites at trace levels.The electrochemical behaviour of a variety of azo compounds has been investigated over the years, and a brief survey of the literature on monoazo compounds alone illustrates the complexity of the electrode reactions involved. Issa et aL6 studied the polarographic reduction of some 4-hydroxy monoazo compounds containing different substituents, and found that reduction proceeded in acidic and alkaline solutions with the consumption of four and two electrons, respectively. In the same study the authors reported that 4-hydroxyazobenzene and azo compounds with weak donor or acceptor groups were capable of catalysing the reduction of H+, this process producing a wave that gave the appearance of a maximum. Similar results were obtained by other In contrast, Solochrome mordant dyes were found by Malik and GuptalO to undergo reduction with the transfer of two electrons in both acidic and alkaline solutions.These dyes produced one wave in acidic solution, but in some instances two waves developed in alkaline media. It has been shown, in an investigation on some para-substituted azobenzenes, that electron-accepting sub- stituents promote reduction to hydrazo derivatives, whereas electron donors drive the reaction partially or totally to the corresponding amines.ll Florence and co-workerslL1* have contributed a number of publications on the study of azo compounds. They suggested that certain species formed unstable hydrazo intermediates, and over-all polarographic n values of 4 were obtained.This paper is concerned with a polarographic study of the four azo dyes CI Direct Orange 34 (I), CI Acid Red 73 (11), CI Direct Blue 84 (111) and CI Direct Red 80 (IV) over the pH range Mechanisms of reduction are postulated on the basis of controlled-potential electrolysis at a large mercury pool and cyclic voltammetry as additional electroanalytical techniques. Optimum pH values ior the determination of the azo dyes by differential-pulse polarography (DPP) are selected and the appropriate linear ranges deduced from calibration graphs. 929 on compounds closely related to those mentioned above. 1-13.930 HART AND FRANKLIN SMYTH POLAROGRAPHIC BEHAVIOUIR AND A%dySt, VOz. 105 Experimental Reagents and Chemicals 0.1 N sulphuric acid and 0.1 N sodium hydroxide prepared in distilled water.Britton - Robinson buffers of pH 2-12 were used and the pH range was extended with S 0 3 - e N = N o N H 2 CI Direct orange 34 (I) C I Acid red 73 ( 1 1 ) OH OH OH OH SO3- SO3- SO3- CI Direct blue 84 ( I l l ) SO3- 0 1 S C ) - O N = N d N = N h so3- \ H N-.C-l 11 I , Mirror image I CI Direct red 80 ( I V ) All four dyes were recrystallised twice from ethanol - water (1 + 1) after which stock solutions were prepared by dissolving the appropriate amounts un distilled water to give concentrations of approximately 5 x Instrumentation All pH measurements were made using an EIL, Model 23A, pH meter, incorporating a glass indicator electrode, saturated calomel reference electrode and a ternperature compensator.Polarography was carried out with a PAR, Model 174A, polarographic analyser operated in the differential pulse mode, and polarograms were recorded on an Advance HR2000 X - Y recorder. A three-electrode system was used for polarography and consisted of a saturated calomel reference electrode, a platinum counter electrode and dropping-mercury electrode (D.M.E.). The dropping-mercury electrode used had the following cha.rac1.eristics: outflow velocity, m = 1.57 mg s-l; drop time, t = 4.3 s at the potential of the saturaked calomel electrode and at a mercury pressure of h = 68 cm in 1 M potassium chloride solution. For the analyser the controlled drop time was 0.5 s (sampled d.c. polarography), 1.0 s (DPP) with a modulation amplitude of 50-100 mV, scan speed 5 mV s-l and low-pass filtler 0.3 s.Cyclic voltammetry was performed with the aid of a PAR, Model 9323, hanging mercury-drop electrode. Con- trolled-potential electrolysis was carried out with the PAR 174A polarographic analyser and a cell containing a large mercury pool. Experimental Techniques M solutions of the dyes in the pH range 1-13. Blanks were obtained by recording polarograms of the appro- priate buffer solutions only, under the same conditions as used for sample solutions. All solutions were de-aerated for 5 min with oxygen-free nitrogen prior to polarography. From the polarographic data obtained by this technique graphs of ilim. v e m a pH and E, versus pH were constructed. M. Sampled d.c. polarography was performed on approximately 5 xOctober, 1980 ANALYSIS OF SOME AZO DYES OF BIOLOGICAL SIGNIFICANCE 93 1 The nature of the electrode process was determined by recording the d.c.polarogram a t varying heights of the mercury column in the appropriate buffer, and graphs of ilim, versus htcorr, were constructed. M solutions of the dyes I-IV in 0.1 N sulphuric acid a t potentials on the plateau of the most negative d.c. polarographic wave. Electrolysis was continued until the current decayed to zero and the electrolysed solution was diluted to give a final concentration of 5 x This final solution was then submitted to spectral and d.c. polarographic analysis using the conditions given previously. Differential-pulse polarography was performed on the four azo dyes I-IV a t the same concentration and using the same pH range as for sampled d.c.polarography. Differential- pulse polarograms of buffer solutions only were recorded using the same conditions. From these polarographic data the optimum pH for determination of the dyes was chosen, and this was used in the construction of calibration graphs. M using a scan rate of 50 mV s-l. Controlled-potential electrolysis was performed on 5 x M. Cyclic voltammetry was performed on some of the dyes a t a concentration of 5 x Results and Discussion CI Direct Orange 34 (I) The variations of iIi,,,. versus pH and E, versus pH show that the dye was reduced in one main wave, i,, which decreased in height a t pH > 5. The height of this wave continued to decrease with increasing pH, then became constant in the pH range 11-13. A small pre-wave, i,, also appeared on the i versus E curves in the pH range 4-13, the height of which remained independent of pH throughout the whole pH range.The E, versus pH relationship of i, in the pH range 1-6 was E+ = 0.06 - 0.09pH. A break occurred on the E+ versus pH plot at pH 6, which is probably a pK, value associated with the protonation of a nitrogen atom in I. Logarithmic analysis [i.e., ED.M.E. versus log i/(id-i)] of the wave i, in 0.1 N sulphuric acid yielded an ma, value of 1.55 and a 9 value of 2.3 for the rate-determining step evaluated from the equation dE,,, - - 0.0599 -- dPH un The variation of limiting current with mercury pressure was determined in 0.1 N sulphuric acid and wave i, showed a linear relationship of ilim. versw ht,,,,. indicating a diffusion- controlled current.When the drop time decreased from 18.2 to 7.0 s the half-wave potential shifted to more negative potentials by 10 mV, suggesting that the electrode reaction was not totally irreversible. At the end of electrolysis the colour and polarographic wave of the compound had both dis- appeared, and ultraviolet - visible spectral analysis of the electrolysed solution showed that the band a t 420 nm had disappeared and a new band had appeared at 300 nm. This suggests that the dye had been reduced in strong acid (pH 1.0) to give a mixture of amines. The mechanism of reduction of this dye (I) in acidic media (pH 1-6) would appear to involve formation of the hydrazo compound, which can then undergo protonation. This two-electron step (E‘) is then followed by another one of similar magnitude (E”) to produce a mixture of 4-aminobenzenesulphonic acid and 1,4-diaminobenzene.In alkaline media the reduction reaction is stopped a t the hydrazo stage, which causes the wave height to decrease by half. The pre-wave i, is believed to be due to product adsorption and it disappeared with the addition of 50% methanol to the buffers a t pH > 7.0. The mechanism of reduction can be illustrated as shown on p. 932. GI Acid Red 73 (11) that I1 is reduced in one main wave (i,) in the pH range 1-4. was Ea = 0.1 - 0.077pH. p value of 2.73 for the rate-determining step, Electrolysis a t a mercury pool in stirred solution was carried out as described earlier. The polarographic waves recorded for CI Acid Red 73 (11) over the pH range 2-12 show The E+ versus pH relationship Logarithmic analysis a t pH 2 yielded an a%, value of 2.0 and a932 HART AND FRANKLIN SMYTH : POLAROGRAPHIC BEHAYIOUR AND Analyst, VOl.105 The variation of ilim, veYsus hacorr, was determined in 0.1 N sulphuric acid and was found to be linear, indicating that the electrode process was diffusion controlled. When the drop time decreased from 16.8 to 6.8 s the Eh value did not shift significantly, which indicates that the rate-determining step at pH 1.0 is reversible. Electrolysis was performed at an applied potential of - 0.2 V f l x 40 min in a supporting electrolyte of 0.1 N sulphuric acid. At the end of the electrolysis period the red colour and polarographic wave (il) had disappeared, and no new waves were observed in the available potential region.Ultraviolet - visible spectral analysis of a portion of this electrolysed solution showed that the band at 513 nm had disappeared completely and the other bands at 346, 330(s) and 245 nm had moved towards the blue end of the spectrum. I H' E" 2e.H' 1 S O 3 - 0 N H 2 t N H z c , F N H 2 Y S It is expected that the azo group (-N = N-) would be first to reduce in a step E', which can be deduced from the work of Florence and co-workers, who found that l-phenylazo-2- naphthol was reduced a t considerably more negative potentials than azobenzene. This is likely to be a reversible step involving 2e and 2H+, followed by protonation of the newly formed hydrazo group and subsequent 4e reduction (E") of the other azo group (-N = N-) to a mixture of amines.The species 4-aminohydrazobenzene can reduce in acidic media to the corresponding amines (E"'). As has been suggested by Florence and co-workers, the rate-determining step is protonation of the hydrazo group, which occurs very rapidly in the pH range 1-4 and thus processes E', E" and E"' occur in one Be step (il). In the pH range 6.0-12.0 the main wave i, is replaced by two waves i, and i, (the latter at considerably more negative potentials) in the ratio 1: 2 with respect to their limiting currents. In this pH range, protonation of the hydrazo compound is difficult and thus the over-all electrochemical process occurs in two steps, involving 2e and 4e, respectively, as shown in the scheme. M (pH 7.0) showed that the steps corresponding to i, and i, are irreversible. In more alkaline media and coinciding with ionisation of the hydroxy moiety (pK, = 1 l.2),19 the polarographic behaviour of I1 changes again in that three waves occur.This could possibly be explained by disproportionation reactions to product 3 such as quinone hydrazones that are reduced in a different manner from azo-containing compounds. CI Direct Blue 84 (111) auersus pH relationship at pH < 2. a P Cyclic voltammetry of CI Acid Red 73 at 5 x In buffer solutions of pH < 6 one wave only was observed, which showed a change in E, This probably arises as a consequence of ionisation of aOctober, 1980 933 sulphonic acid group and also affected absorption bands in the ultraviolet spectrum around the same pH value. The E , versus pH relationship in the pH range 1-2 was The variation of ilim.versus Mcorr, for I11 in 0.1 N sulphuric acid was linear and showed the current to be diffusion controlled. A change in drop time from 16.2 to 7.1 s caused the E , to shift to more negative potentials by 45 mV, which indicates that the electrode process is irreversible. Electrolysis was performed at a potential of -0.6 V for 40 min. At the end of the electrolysis time the visible band and ultraviolet bands a t 290 and 350 nm had dis- appeared, and a low-intensity band appeared a t 270 nm upon which two shoulders were ANALYSIS OF SOME AZO DYES OF BIOLOGICAL SIGNIFICASCE E , 3 -0.43 - 0.04pH. k Hi934 HART AND FRANKLIN SMYTH : POLAROGRAPHIC BEHAVIOUR AND Analyst, VOl. 105 observed at 240 and 220 nm.The polarogram obtained after electrolysis at -0.6 V did not show any waves in the potential range 0 to -1.0 V, which again suggests the formation of amine derivatives upon electrolysis in strong acid (pH 1.0). Logarithmic analysis of the polarographic waves obtained in Britton - Robinson buffers of pH 3.0 and 4.0 yielded cma values of 1.3 and 1.6 and p values of 1.5 and 1.9, respectively. This suggests that in acidic media the rate-determining step involves 2e and 2H + and, unlike species I and 11, that protonation of the hydrazo intermediate is not rate determining. The E, value of I11 in acidic solution is far more negative than that of either I or 11, which illustrates both the effect of hydrogen bonding between the azo group and two ortho-hydroxy groups and inductive effects.The mechanism of reduction of CI Direct Blue 84 in the pH range 1-6 can be represented by the scheme shown below, in which hydrogen bonding has been omitted. OH OH OF The changes in Ea veysus pH and ilim. versus pH graphs at pH 6 suggest an acid - base equilibrium with pKa = 6.0. This is in reasonable agreement with the value obtained spectrophotometrically, i.e., pK, = 5.2, which was considered to reflect ionisation of the hydroxy moieties at the naphthalene 8-positions.19 Between pH 6 and 12 a new wave, i,, appeared on the current - potential curves, which increased in height with increasing pH while the height of the main wave i, decreased. At pH 12 the wave i, reached a height equivalent to a four-electron reduction. One explanation for the appearance of wave i, is the reduction of a quinone hydrazone species (V), which is formed after the 8-hydroxynaphthalene group has ionised (i.e,, at pH 6.0).As the pH increased the formation of species V is facilitated, until at p H 12 maximum electro- reduction is attained. The equilibrium existing in solution at pH m pK, can be represented as shown on p. 935. Tautomerism similar to that shown has been reported to exist for other dyes of closely related structure in aqueous solutions.October, 1980 CI Direct Red 80 (IV) The variations of ilim. versus pH and Eb versus pH for CI Direct Red 80 show two waves, i, and i,, with relative heights of 1:4, observed on the current - potential curves between pH 1 and 4 with E , = $0.1 - 0.066pH and E , = O-O.O66pH, respectively.The variations of limiting current and half-wave potential with mercury pressure were examined in 0.1 N sulphuric acid. Both i, and i, showed linear relationships in the graphs of ilim. versus h+,,,,., indicating diffusion-controlled electrode reactions. When the drop time decreased from 18.7 to 6.8 s the Ei value of i, and i, shifted by 10 mV, which suggests that the electrode processes giving rise to these waves are not totally irreversible. ANALYSIS OF SOME A 2 0 DYES OF BIOLOGICAL SIGNIFICANCE 935 f i N = N w N = N f i 0- OH OH 0- SO3- SO3- SO3- SO3- ( I l l ) / / It (V) Electrolysis was carried out a t a stirred mercury pool held a t a potential of -0.2 V for 40 min. At the end of the electrolysis period the red colour of the solution had disappeared and ultraviolet - visible spectral analysis showed that the visible bands a t 420, 520 and 545 nm had completely disappeared and a new band of low intensity appeared at 340 nm.All the bands in the ultraviolet region showed blue shifts after electrolysis. Polarographic analysis was carried out on the same electrolysed solution and no waves appeared on the current - potential curve in the potential range s0.160 to - 1.080 V. This indicates that reduction of IV in strong acid (pH 1.0) proceeds with the consumption of 16 electrons, giving the corresponding amines. Logarithmic analysis of the wave i, at pH 2 gave an a.n, value of 2.3 and a $ value of 2.7. The large an, value was considered to reflect the simultaneous addition of two electrons to each of the azo end-groups in the rate-determining step.However, although the p value apparently indicates the addition of three protons in the reaction, it was believed to involve six protons. This may be explained by considering that the linear portion of the E , veYws pH graph of i, represents two identical graphs superimposed on each other. This is not unreasonable when it is noticed that the molecule is symmetrical and that the azo end-groups are in identical environments. Thus the Ea values associated with the respective reductions of these two moieties might be expected to be so close that they would be indistinguishable. The an, and p values obtained for wave i, at the same pH were 2.7 and 2.8, respectively, which again suggests that four electrons and six protons are involved in the rate-determining step.The mechanism of the reduction of CI Direct Red 80 in the pH range 1-4 may be rep- resented as shown in Fig. 1. The wave i, arises from process E' and i, from E", E"' and E"", which occur simultaneously. In the pH range 6-12 process E"" no longer occurs because protonation of the hydrazo end-groups is difficult. This is reflected by the decrease in the height of i, to a value equivalent to eight electrons while i, remains independent of pH through the pH range 1-12. At pH greater than 12 both i, and i, decrease in height and two new waves, i, and i,, appear on the current - potential curves. This indicates an acid - base equilibrium with pK, = 12-13 and is associated with ionisation of hydroxy groups.lg936 HART AND FRANKLIN SMYTH : POLAROGRAPHIC BEHAVIOU;R AND Afialyst, Vol.105 M solution of the dye in 0.1 x sodium hydroxide solution and on scanning in a negative direction a peak was observed at -0.700 V and another broader peak at -0.860 V. On reverse scanning only one peak was observed, at -0.680 V, which was believed to arise from re-oxidation of the hydrazo derivative (VI) to the parent dye. Cyclic voltammetry was performed on a 5 x Variation of ip with Concentration and Differentiation of I-IV by Differential-pulse Polarography The optimum pH for analysis of the azo dyes I-IV was chosen from a consideration of both the magnitude and shape of the differential-pulse polarographic peaks, recorded at con- centrations of approximately 5 x M in the pH range 1-13. OH s o 3 - e ; ; : e N = N b H \ / - , N-C- I Mirror image SO3- E"" 4e.2HC I so3- 2 N H 2 d NH:, f 2 SO3- +NH2 Fig.1. Reduction scheme for CI Direct Red 80. All four dyes showed the best defined peaks for quantitative analysis in acidic media, e.g., the variation of log[peak current (PA)] veysus log[concentration (pmol 1-l)] for CI Direct Red 80 at pH 1.0 shows linearity of response over the concentration range 5 x 10-'-4 x M. Above 4 x M the graph begins to level off, probably as the reijult of adsorption processes. Table I summarises the differential-pulse polarographic data for the four dyes obtained at the selected pH values. It can be seen that the dyes showed linearity of response to varyingOctober, 1980 ANALYSIS OF SOME AZO DYES OF BIOLOGICAL SIGNIFICANCE 937 TABLE I DIFFERENTIAL-PULSE POLAROGRAPHIC DATA FOR AZO DYES Supporting Response Range of concentration Compound electrolyte Ei; V factor/@.Gmol-'l for linear responselhl CI Direct Orange 34 , . Britton . Robinson buffer, pH 4.0 - 0.280 It 0.005 3.6 x 5 X 10-'-3 X lo-' CI Acid Red 73 . , , . Britton - Robinson buffer, pH 4.0 - 0.220 i 0.005 1.38 x lo-* 5 x lo-'-5 x CI Direct Blue 84 . . Britton . Robinson buffer, pH 3.0 - 0.605 It 0.005 2.05 x 1 X 10-'-5 X CI Direct Red 80 , . . . 0.1 N H,SO, -0,010 + 0.005 1.15 x lo-* 5 x 10-'-4 x lo-& extents, but all gave a linear response in the lower concentration range. The response factor was determined from the slope of the i, (PA) veYsus concentration (kmol 1-l) graphs, where concentrations in the range 5 x 10-7-5 x Table I shows that all four dyes I-IV can be determined a t the trace level by using differential-pulse polarography.A further aspect of the differential-pulse polarographic technique is that it offers the possibility of differentiating and identifying the azo dyes in alkaline media. This is illustrated in Fig. 2, which shows the unique profiles obtained a t concentrations of approximately 5 x M were investigated. M in 0.1 N sodium hydroxide solution. n, Potential Fig. 2. Differential-pulse polarograms of (A) CI Direct Blue 84, (B) CI Direct Red 80, (C) CI Direct Initial potential, -0.2 V. Supporting electrolyte, 0.1 N sodium Orange 34 and (D) CI Acid Red 73. hydroxide solution. Conclusion The polarographic behaviour in acidic solution suggested that all four azo dyes were reduced to the corresponding amines.However, at pH > 6 end-groups containing sub- stituted azobenzenes in CI Direct Orange 34, CI Acid Red 73 and CI Direct Red 80 were reduced only to the hydrazo derivatives, whereas the remaining azo groups in the red dyes reduced to the amines. The polarographic behaviour of CI Direct Blue 84 in the same pH region was different to that of the other three dyes, which was considered to reflect the presence of tautomerism. The optimum conditions for quantitative analysis of the four azo dyes were in acidic media, where as alkaline solutions proved more suitable for qualitative purposes.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. HART AND FRANKLIN SMYTH References Games, L. M., and Hites, R. A., Alzal. Chew., 1977, 49, 1433. Donaldson, E., DIFS, Belfast, personal communication. Howe, J. R., Lab. Pract., 1975, 24, 457. Miller, J. A,, and Miller, E. C., Adv. Cancer Res., 1953, 1, 339. Cilento, G., Miller, E. C., and Miller, J. A,, J . Am. Chem. Soc., 1956, 78, 1718. Issa, I. M., Issa, R. M., Temerk, Y . M., and Mahmoud, M. R., Electrochim. Acta, 1973, 18, 139. Holleck, L., and Holleck, G., Naturwissenschaften, 1964, 51, 433. Castor, C. R., and Saylor, J. H., J . Am. Chem. Soc., 1953, 75, 1427. Shams El-din, A. M., J . Electroanal. Chem. Interfacial Electrochem., 1969, 21, 377. Malik, W. V., and Gupta, P. N., J . Electvoanal. C h e w Interfacial Electrochem., 1974, 54, 417. Jannakoudakis, D., Kokkinidis, G., and Mauridis, P. G., J . Chim. Phys. Chim. B i d , 1976, 73, 872. Florence, T. M., and Farrar, Y . J., Aust. J . Chem., 1964, 17, 1085. Florence, T. M., and Aylward, G. H., Aust. J . Chem., 1962, 15, 65. Florence, T. M., and Aylward, G. H., Aust. J . Chem., 1962, 15, 416. Florence, T. M., Aust. J . Chem., 1965, 18, 609. Florence, T. M., and Belew, W. L., J . Electroanal. Chem., 1969, 21, 157. Florence, T. M., Johnson, D. A., and Batley, G. E., J . Electroanal. Chem., 1974, 50, 113. Florence, T. M., J . Electroanal. Chem. Interfacial Electrochem., 1974, 52, 115. Hart, J. P., and Franklin Smyth, W., Spectrochim. Acta, 1980, 36A, 279. Received May 2nd, 1979 Accepted June Znd, 1980