ANALYST, JANUARY 1993, VOL. 118 101 Spectrophotometric Titration of Some Thiazine Dyes With Iron(ii) in Buffer Medium in the Presence of Oxalate K. Vijaya Raju and G. Bangar Raju Department of Engineering Chemistry, Andhra University, Visakhapatnam 530 003, India An accurate and convenient spectrophotometric titration method was developed for the determination of microgram amounts of seven thiazine dyes employing iron(l1) as a reductant in buffer medium (optimum pH 3.7-4.7) in the presence of sodium oxalate (optimum concentration 0.06-0.12 moll-1). All the dyes are rapidly and quantitatively reduced to their colourless leuco-bases in a two-electron reduction with iron(l1). The redox potentials of the iron(lll)-iron(ll) couple in buffer media of different pH and oxalate ion concentrations were measured.The function of pH and oxalate ion in the reaction medium is discussed. Keywords: Spectrophotometric titration; thiazine dye; iron(//) reductant; buffer medium; oxalate Thiazine dyes find numerous applications such as in textile technology,' medicine and biology2 and as redox indicators.34 Surprisingly, only a few titrimetric methods, involving the use of titanium(ni) chloride2.7 and iron(ii)8+9 in the presence of phosphoric acid as reductants, have so far been reported for the assay of these dyes. For the determination of a few dyes (Methylene Bluelo-17 and Thionine14-17) among the group, however, some other reductants [tin(ii) ,lo vanadium(1r) ,I1 chromium(i1) ,I2 molybdenum(iii)13 and ascorbic acid141 and a few oxidants [cerium(iv)15,16 and chloramine-T17] have also been used.All these methods suffer from one disadvantage or another. For example, the preparation and preservation of some of the reductant solutions2~7~10--'3 is tedious and the titrations using them need be carried out at elevated tempera- tures. The high concentration of orthophosphoric acid required in the iron(i1) method makes the medium viscous and the method expensive; in addition, a correction factor is necessary for calculating the dye content. In the oxidimetric methods reported it is difficult to ascertain the end-product of a dye and hence the stoichiometry of the reaction. This paper describes a spectrophotometric titration method for the determination of thiazine dyes in which the dye solutions are taken in a buffer medium containing oxalate and titrated with iron(i1).The method obviates all the disadvan- tages of the earlier methods. Further, the reagents of the reaction medium are relatively inexpensive and available in high purity. As most of the commercial dyes are generally found impure, assay methods for dye samples are of great importance. Even the manufacturers of the dyes (except a few) do not furnish the percentage dye contents of the samples. Experimental Reagents All solutions were prepared with distilled water and all chemicals were of analytical-reagent grade unless stated otherwise. Zron(rr) solution , 0.01 rnol 1 - 1. Prepared from ammonium iron(1i) sulfate hexahydrate in 0.01 mol 1-1 sulfuric acid and standardized18 against a standard solution of potassium dichromate. A portion was diluted further to 5.0 X 10-4 rnol 1-1 (in 5.0 x 10-4 rnol 1-1 sulfuric acid).Iron(ir1) solution, 0.01 rnol 1-1. Prepared from ammonium iron(i1i) sulfate dodecahydrate in 0.01 moll-' sulfuric acid and standardized18 by reduction with tin(1r) and then titration with a standard solution of dichromate. Titanium (111) chloride solution , 0.01 mol I- 1. Prepared from 15% m/v titanium chloride in 2 moll-' hydrochloric acid and standardized19 with a standard solution of dichromate. Sodium oxalate sohtion, 0.02 rnol 1-1. Prepared from disodium oxalate in distilled water. Buffer solutions. Various buffers of the desired pH (1-5) were obtained20 by mixing suitable volumes of 1 rnol 1-1 hydrochloric acid and 1 rnol 1-1 sodium acetate in a total volume of 50 ml.Purification and solutions of the dyes. The purity of the dye samples was tested qualitatively by thin-layer chromatography with acetic acid-water (1 + 1) as the mobile phase. The dyes, Azure-A (AZA), Azure-B (AZB) and Azure-C (AZC) (Gurr) were found to be associated with some coloured impurities. These dyes were rendered free from such impuri- ties by silica gel column chromatography as follows. The coloured impurities were first removed by elution with chloroform. The dye was then eluted with chloroform- methanol (with a gradual increase in the concentration of the former) and finally with methanol (the dye being highly polar) to minimize the recovery loss. The purified dyes were used for analysis. The other dyes, Thionine (THN) (Riedel-de Haen), Methy- lene Blue (MB), New Methylene Blue (NMB) (both from Merck) and Toludine Blue (TLB) (Gurr) were found to be free from coloured or visible impurities.The colourless impurities are generally due to additives1 such as sodium chloride and sodium sulfate used in the preparation of the dyes1 and do not interfere in the proposed method. Aqueous 0.05% solutions of all the dyes (50 mg in 100 ml) were prepared and standardized against titanium(iI1) chloride.2-7 From these standard dye solutions 5.0 X 10-5 moll-' solutions were prepared by suitable solution. Leuco-dye solutions. Solutions of the leuco-dyes were prepared in 50 ml calibrated flasks by taking a known aliquot (5.0 ml) of the dye solution (5.0 x 10-4 rnol 1-1) in a buffer (pH =4.20) medium containing 0.06 rnol 1-1 sodium oxalate and adding a 50-fold excess of iron(n) (5.0 ml of 0.03 moll-' solution) and finally diluting to the mark.Apparatus A Shimadzu (UV 140.02) double-beam spectrophotometer with optically matched glass cells of 1 cm pathlength was used to record the absorption spectra. An optical glass cell of 3 cm pathlength (3 x 5 x 6 cm) was employed for the spectro- photometric titrations. As the cell holder for such a glass cell to fit on to the cell compartment is not provided with the spectrophotometer, a102 ANALYST, JANUARY 1993. VOL. 118 Table 1 Spectrophotometric titration of some thiazine dyes with iron( 11) Dye* AZA AZB AZC MB NMB THN TLB Referencc met hodz-7 73.0 116.7 145.9 218.9 87.6 140.0 175.1 262.6 69.4 111.0 138.8 208.2 93.5 149.5 186.9 280.3 104.0 166.4 208.0 312.0 71.8 114.9 143.6 215.4 76.4 122.3 152.9 220.3 Proposed method 73.5 116.1 146.5 219.5 88.2 139.4 175.8 263.4 70.0 110.5 139.3 208.8 94.1 148.8 187.6 281 .0 103.1 167.6 208.7 311.1 71.2 115.7 143.0 214.8 77.0 121.7 153.5 229.9 Found/pgl Relative standard deviation 0.5 0.4 0.4 0.3 0.6 0.4 0.3 0.2 0.5 0.4 0.3 0.2 0.4 0.3 0.3 0.2 0.4 0.3 0.2 0.2 0.6 0.5 0.4 0.3 0.5 0.4 0.3 0.3 (%) Conditional redox pot enti al/ Assay mV (%)$ ( k 5 m V ) 88.9-t-0.4 411 87.5 -1-0.3 407 81.3 k 0.3 403 X8.4f0.5 425 90.5 -1-0.3 401 92.5 k0.3 398 90.0 k 0.3 418 contents were not labelled on the samples * Percentage dye - - supplied.t Average of six determinations. $ Confidence interval for the mean of six determinations (95% probability). few modifications to the cell compartment were made as described earlier21 in order to carry out the titrations.A digital potentiometer and a digital pH meter were used for potential and pH measurements, respectively. Procedure To a known aliquot (5-15 ml) of the dye solution ( 5 x 10-5 rnol 1-1) placed in the titration cell, 10.0 ml of sodium acetate (1 rnol I - l ) , 8.0 ml of hydrochloric acid (1 moll-1) and about 15 ml of sodium oxalate (0.2 rnol 1-1) solutions were added and diluted to SO ml (observed pH 4.20 k 0.01). The titration cell was then placed in position in the spectrophotometer, which was initially set to the desired wavelength (see Absorption Spectra of the Dyes) and at zero absorbance with respect to a blank. Purified nitrogen was bubbled through the reaction mixture for about 2 min to homogenize the solution and to expel any dissolved oxygen present and the absorbance was noted.The iron(i1) solution (5.0 X 10-4 rnol 1-1) was added in increments. After adding each portion of the titrant, the solution in the cell was stirred for about 30 s by passing nitrogen, which incidentally served as the inert atmosphere needed in the titration (to prevent the aerial oxidation of the Zeuco-dye or reduced product). The stirring was then stopped and the absorbance noted. The titration was continued in this way until the absorbance became almost constant. A plot of absorbance (corrected to dilution) versus volume of titrant added consists of two straight lines and their point of intersection corresponds to the equivalence point.Some typical results and assay values obtained are given in Table 1. 1.4 1 .o u m 5 s 2 0.6 0.2 n 400 480 560 640 720 800 Wavelengthlnm H '-/% 2H+ + 2e- f i Thiazine dye leuco-Base Dye CINo. R1 R2 AZA 52005 NH2 N(CH3)Z AZB 52010 NHCH3 N(CH3)2 MB 52015 N(CHS)zN(CH3)2 THN 52000 NH2 NH2 Fig. 1 Absorption spectra of A, AZA; B, AZB; C, MB; and D, THN in buffer medium (pH ~ 4 . 2 ) in the presence of 0.06 mol 1-I sodium oxalate. Concentration of each dye = 3.5 x mol I-' Results and Discussion Absorption Spectra of the Dyes In order to select a suitable wavelength for the spectropho- tometric titration of each dye, the absorption spectra of the dyes, their feuco-bases, iron(I1) and iron(II1) in media of various pH (range 4-5) and oxalate ion concentrations were recorded over the range 400-800 nm.The iron solutions and the leuco-bases have negligible absorbance in the visible region. The spectra of all the dyes (Figs. 1 and 2) were found to be independent of the pH of the medium (4-5) and the concentration of sodium oxalate (0.01-0.12 rnol 1-1). The optimum wavelengths for the spectrophotometric titrations are (Figs. 1 and 2) 600 nm for THN, 620 nm for TLB and NMB, 640 nm for AZA and AZC, 660 nm for AZB and 680 nm for MB. Beer's Law and Stability of the Dye Solutions Beer's law was found to be obeyed from about 1.5 to 4.5 pg ml-1 (3 cm pathlength) for THN, AZA, AZC and TLB. and from 2.0 to 6.0 pg ml-1 for AZB, MB and NMB. A spectrophotometric study of the stability of the dye solutions and leuco-bases under the titration conditions of the proce- dure (pH -4.2 in the presence of 0.06 moll-' sodium oxalate) revealed that all the dye solutions were stable for 3 d.However, the Zeuco-dye solutions were found to be stable towards oxygen in air for only 3 h even in the presence of a 50-fold excess of iron(r1). Effect of pH and Oxalate Ion Concentrations and Stoichiometry of the Reaction Preliminary investigations of these titrations revealed that of various buffers tried, the hydrochloric acid-sodium acetate buffer appears to be the best. Further, the pH of the buffer must be in the range 3.7-4.7 and the concentration of oxalateANALYST, JANUARY 1993. VOL. 118 103 Wavelengthhm H Thiazine dye leuco- Ba se Dye CINo. R' R2 R3 R4 TLB 52040 NH2 CH3 H N(CHd2 NMB 52030 NHC2H5 CH3 CH3 NHC2H5 AZC 52002 NH2 H H NHCH3 Fig.2 Absorption spectra of A, AZA; B, TLB; and C, NMB in buffer medium (pH -4.2) in the prcscnce of 0.06 mol 1-1 sodium oxalatc. Concentration of each dyc = 3.5 x 10-5 mol 1-1 500 400 W I z v, $ 300 > > E 1 (D .- c 5 200 * n 50 I I I I 0 0.04 0.08 0.12 Concentration of sodium oxalate/mol I-' Fig. 3 Formal redox potentials of the iron(ii)-iron(iii) couple in a buffer medium ( H =4.0-4.3) at different concentrations of sodium oxalate. lrron(ii,P = [iron(iii)l= 0.001 moll-1; temperature = 28 * 0.1 "C 0.06 mol 1-1 or above for rapid reduction and satisfactory titration of the dyes with iron(ii). All these dyes are reduced by iron(I1) to their corresponding leuco-bases in a two-electron reduction.2.7 Redox Potentials As early as in 1931, Michaelis and Friedheirn2' reported that the potential of the iron system decreases in a buffer medium containing oxalate.However, the use of iron(i1) as a useful reductimetric titrant in such a medium has recently been introduced in our laboratories.23J4 The redox potentials of the iron(rr1)-iron(ii) system at different pH and oxalate ion concentrations have not pre- viously been measured. Hence these potential values were 400 W $ 320 v, $ > 240 E 1 .- c 160 n 80 0.5 1.5 2.5 3.5 4.5 5.5 PH Fig. 4 Formal redox potentials of the iron(ii)-iron(iii) couple in buffer mcdia of various H (0.8-5.3) in the prescncc of 0.06 mol 1-1 sodium oxalate. [Iron(ir)r= [iron(iii)] = 0.001 moll-1; temperature = 28 k 0.1"C Table 2 pH of the buffer (3.95 k 0.01)* containing equimolar amounts of iron(iii) and iron(ir) at different concentrations of sodium oxalatc.[Fe"] = [Fe"'] = 0.001 mol 1-I. temperaturc = 29 C O.l"C, total dilution = 50 ml concentration of pH Concentration o f pH 0.0 3.95 0.06 4.20 0.01 4.02 0.08 4.27 0.02 4.07 0.10 4.32 0.04 4.13 0.12 4.33 oxalate/mol1-1 f 0.01 oxalate/mol1-~ Zk 0.01 * pH of the buffer under titration conditions of the procedure in the absence of oxalatc. determined adopting the method of Rao and Dikshitulu25 to explain the redox reaction between the dyes and iron(ii). The results for fixed pH and varied sodium oxalate are shown in Fig. 3 and those for fixed sodium oxalate and varied pH in Fig. 4. The acid contribution from the iron salt solution was taken into account when adjusting the pH of the solution.The pH of the buffer was found to increase to a small extent with increase in oxalate ion concentration (Table 2). The potentials are reproducible to within k5 mV and the pH of the medium was checked experimentally before each potential measurement. The redox potentials of the dye couple (ox.dye)-(red.dye) could not be determined in the same way because of the difficulties in the preparation of the leuco-dyes, which are highly sensitive to atmospheric oxygen. However, the condi- tional redox potentials of the dye couples were determined by measuring the potential of the dye solution that had been exactly 50% titrated with iron(r1) under a nitrogen atmosphere and are given in Table 1 . From a comparison of the conditional redox potentials of the dye couples (Table 1) and the redox potentials of the iron system (Figs.3 and 4), it can be seen that at a pH of about 4.20 (pH observed under the titration conditions) in the presence of 0.06 mol I-' sodium oxalate there is sufficient difference (290-320 mV) between the redox potential of the iron system (105 mV) and the dye systems (398425 mV) to bring about rapid reduction of the dyes with iron(ir). The amount of sulfuric acid present in the titrant, iron(Ii), is small (overall acidity 5.0 x 10-4 mol 1-1) and was found to have almost no effect on the buffer (pH 4.2 2 0.01) during the titration. It is well known that the redox potential of the iron(irr)- iron(1r) system decreases in the presence of certain complexing ions ( e . g . , fluoride, pyrophosphate, phosphoric acid, ethyl- enediaminetetraacetic acid) that bind iron(1Ii) in the form of a complex much more strongly than they do iron(1r) and enhance the reducing ability of iron(1r).In the present instance also, the decrease in potential of the iron system by oxalate is104 no doubt due to the formation of a stronger complex between iron(m) and oxalate (log p1 = 7.53, log 8 2 = 13.64 and log p3 = 18.49)26 than that between iron(ii) and oxalate (log p1 = 3.05, log p2 = 5.15).26 However, the effect of pH on the oxalato complex of iron(rr1) has to be taken into account, as the oxalate anion is readily protonated (pKl = 1.3, pK2 = 4.3).27 This mainly accounts for the change in the iron(iii)-iron(rI) potential as a function of pH and of the overall oxalate ion concentration of the medium.At a fixed pH, the potential will decrease (Fig. 3) with increase in oxalate ion concentration, because of the greater extent of complexation, and at fixed oxalate ion concentration the potential will increase (Fig. 4) with decrease in pH, because of the greater protonation of oxalate ion. Nature of the Complex Expected to be Formed Between Iron(m) and Oxalate In a medium of pH ==4.20 (observed pH under the titration conditions of the procedure), about half (0.028 moll-1) of the total oxalate (0.06 mol 1-1) is expected to be present in the form of C2042- ions according to the equation28 K,’ K,” [OX], [C2042-] = (1) [H+]2 + K,’ [H+] + K,’ K,” where K,‘ (5.6 x 10-2) and K,” (5.4 X 10-5) are the first and second dissociation constants27 of oxalic acid and [OX], represents the total oxalate (0.06 mol 1-1) in the medium.Iron(ii1) is tightly and more strongly bound by C2O42- ion (log = 7.53, log p2 = 13.64 and log p3 = 18.49)26 than by HC204- ion (log K = 4.3926 and is capable of forming 1 : 1 (mono) [Fe(C204)]+, 1 : 2 (bis) [Fe(C204)2]- and 1 : 3 (tris) [Fe(C204)3]3- complexes with oxalate ion .29-31 The fraction of the complexes present in the system can be calculated from the equations.32 Ki K2 K3 [c2042- l3 N [MOX~] = where M represents Fell1 and N = 1 + K1[C204~-] + KlK2 [C2042-]2 + KiK2K3 [C2042-]3. K1, K2 and K3 are stepwise stability constants of the complexes (K1 = 3.4 X 107, K2 = 1.3 x 106 and K3 = 7.1 x 104).26 From eqns. (2), only the 1 : 3 complex or trisoxalatoferrate(II1) , [Fe( C204)3]3- , must be predominant at pH 4.20 (on calculation, the fractions of the MOx, MOx2 and MOx3 complexes present in the system are found to be 0.14 x lO-7,0.50 x 10-3 and 0.98, respectively).Attempts to detect the end-point of these titrations using a potentiometric technique did not succeed because of the lack of a clear break in potential at the equivalence point. ANALYST, JANUARY 1993, VOL. 118 Interferences Chloride, sulfate, acetate, perchlorate and nitrate ions do not interfere. However, the nitrite ion interferes. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 References Venkataraman, K., The Chemistry of Synthetic Dyes, Academic Press, New York, 1952, vol. 2, p. 780. Conn, H. J., Biological Strains, Williams and Wilkins, Balti- more, 9th edn., 1977, pp.415430 and 602-605. Bishop, E., Indicators, Pergamon Press, Oxford, 1972, pp. 506-509. Krishna Murthy, N., and Satyanarayan, V., Acta Cienc. Indica. 1977,3, 201. Krishna Murthy, N., Dakshina Murthy, P. M., and Krishna Rao, P. V., Acta Cienc. Indica, 1985, 11, 181. Vijaya Raju, K., Bangarraju, G., and Madhu Gautam, G., J. Znst. Chem. (India), in the press. Knecht, E., and Hibbert, E., New Reduction Methods in Volumetric Analysis, Longmans, London, 1925, p. 101. Rukmini, N., Ph.D. Thesis, Andhra University, 1964. Ramanadham, G. V., Ph.D. Thesis, Andhra University, 1976. Leutweim, F., Fresenius’ 2. Anal. Chem., 1940, 120,233. Banerjee, P. C., J . Indian Chem. SOC., 1942, 19, 35. Tandon, J. P. , and Mehrotra, R. C., Fresenius’ 2. Anal. Chem., 1957,158, 189.Sagi, S. R., and Bosu Babu, T., Talanta, 1976, 23,465. RfiiiEka, E., and Kotoucek, M., Fresenius’ Z. Anal. Chem., 1961, 180, 429. Murali Krishna, U., and Ramanatham, G. V., J. Indian Chem. SOC., 1976, 58, 95. Murali Krishna, U., Ramanatham, G. V., and Kanna Rao, P., Acta Cienc. Indica, 1976, 2, 344. Chermavina, M. S., Tr. Sverdl. Skh. Znst., 1960, 7, 363. Vogel, A. I., A Text Book of Quantitative Inorganic Analysis, Longmans, London, 4th edn., 1978, pp. 360 and 399. Breit, J. E., J. Assoc. Off. Agric. Chem., 1949, 32, 589. Vogel, A. I., A Text Book of Quantitative Inorganic Analysis, Theory and Practice, Longmans, London, 2nd edn., 1951, p. 869. Vijaya Raju, K., Madhu Gautam, G., and Bangarraju, G., Mikrochim. Acta, in the press. Michaelis, L., and Fricdheim, E . , J. Biol. Chem., 1931,91,343. Murthy, N. K., Pulla Rao, Y., and Satyanarayana, V., J. Indian Chem. SOC., 1978, 55,686. Pulla Rao, Y ., Prasad, G. V., and Murthy, N. K., Analyst, 1987, 112, 1777. Rao, G. G.. and Dikshitulu, L. S. A., Talanta, 1962, 9, 715. Smith, R. M., and Martell, A. E., Critical Stability Constants, Plenum Press, New York, 1977, vol. 3, pp. 93 and 94. Haight, G. P., Jr.. and Huber, C. F., J. Am. Chem. SOC., 1976, 98, 14. Day, R. A., Jr., and Underwood, A. L.. Quantitative Analysis, Prentice-Hall, Englewood Cliffs, NJ, 5th edn., 1986, p. 213. Lambling, J . , Bull. SOC. Chim. Fr., 1949, 16, 495. Lingane, J. J . , J. Am. Chem. SOC., 1946, 68, 2448. Bobtelsky, M., Chasson, D., and Klein, S. F., Anal. Chim. Acta, 1953,8,460. Subba Rao, P. V., Krishna Rao, G. S. R., Rama Krishna, K., and Murthy, P. S. N., Indian J . Chem., Sect. A , 1991,30,239. Paper 2104.5736 Received June 11, 1992 Accepted August 26, 1992