Analyst, April, 1971, Vol. 96, pp. 281-285 28 1 A Spectrofluorimetric Method for the Determination of Small Amounts of Sulphate Ion BY LAY HAR TAN AND T. S. WEST (Chemistry Department, Imperial College, London, S. W.7) The enhancement of the fluorescence of the binary complex of zirconium with Calcein blue a t pH 1.9 is used to determine sulphate ion in the range 0.2 to 12 mg (2 to 12 000 p.p.m.). Excitation and fluorescence maxima occur at 350 and 410 nm, respectively. The fluorescence is stabilised immediately and remains unchanged for over 4 hours. Fluoride present in low con- centrations gives rise to high results and must be absent. Other anions that form complexes with or precipitate zirconium, e.g., oxalate, phosphate, tartrate and tungstate, cause low results, but there is a high tolerance towards most cations except iron(II1) and cobalt(I1). THE most widely used methods for the determination of moderately large amounts of sulphate involve precipitation of barium sulphate followed by a gravimetric or titrimetric procedure.Smaller amounts can be measured turbidimetrically or nephelometrically after precipitation. Several indirect spectrophotometricl and fluorimetric2 procedures have also been described, based on the bleaching or quenching action of sulphate ion on some highly coloured or fluorescent metal - dyestuff complexes. Jones and Letham3 have described another novel indirect method based on the use of chloroaminobiphenyl. Recently a kinetochromic pro- cedure has been used to provide a direct absorptiometric method based on the catalytic effect of sulphate ion on the reaction between methylthymol blue and a partially polymerised solution of zirconyl ions.4 A similar fluorimetric procedure has been devised for sulphate ion in which the fluorescent reagent morin is used instead of methylthymol blue.5 Although the last two methods mentioned above are very sensitive they are kinetically controlled and require to be applied under rigidly maintained conditions.In this paper, we describe a direct fluorimetric determination that depends on the enhancement of fluorescence produced by allowing sulphate ion to react with an unpolymerised zirconium solution and the fluorescent reagent Calcein blue. We have recently described a similar reaction for fluoride ion6 based on the zirconium - Calcein blue system and have described the probable mechanism of the reaction resulting from the formation of a ternary complex between the three reactants.The sulphate reaction is much weaker than the fluoride reaction and it was found necessary to use a considerable (100 to 1000 mole ratio) amount of sulphate ion to obtain a linear calibration graph. Nevertheless, although this requirement completely vitiated our efforts to establish the existence of a similar ternary complex involving sulphate ion, zirconium and Calcein blue, we were easily able to establish linear relationships between the increase of fluorescence and the amount of added sulphate. M solution of Calcein blue (curves A and A’, respectively), an exactly formulated 1 : 1 zirconium - Calcein Fig.1 shows the excitation and emission spectra of a 0 SAC and the authors.282 TAN AND WEST: A SPECTROFLUORIMETRIC METHOD FOR [A~alyst, Vol. 96 '1 'IExcitation 60 50 - 40 - 300 350 & 400 \ I I I I I I I I I I I I 400 450 500 Wavelength/nm Fig. 1. Excitation and emission spectra Curves A and A': Curves B and B' : 1 0 - 6 ~ 1 : 1 Calcein blue - zirconium Curves C and C': 10-'JM 1 : 1 Calcein blue - zirconium Excitation spectrum Emission spectrum M Calcein blue + 200-fold excess of sulphate A: emission at 450 nm B: emission at 410 nm C: emission at 410 nm A': excitation at 325 nm B': excitation a t 350 nm C': excitation at 350 nm 70 60 50 40 30 20 10 0 1 2 3 4 5 6 7 lo-* M Calcein blue/ml Fig. 3. Effect of concentration of Calcein blue: curve A, zirconium - Calcium blue - sul- phate [2 ml of lo-* M zirconium(1V) in 3 M hydro- chloric acid + 4 ml of 10-2 M potassium sulphate] ; curve B, zirconium- Calcein blue (as A but without potassium sulphate) ; and curve C, ?effect caused by sulphate ion (B-A) 70L 6orio 50 :: 20 l o t ' 1-8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 1 1 1 1 I I I I 1 1 I PH Fig.2. Effect of pH on sensitivity: curve A, zirconium - Calcein blue - sulphate [2 ml of M Calcein blue + 2 ml of lo-* M zirconium(1V) in 3 M hydrochloric acid + 4 ml of 10-2 M potassium sulphate in 100 ml) ; and curve B, zirconium - Calcein blue (as A but without potassium sulphate) n Fig. 4. Attempted continuous vari- ation graph: broken line, net effect, i.e., corrected curve; n = millilitres of 10-2 M potassium sulphate added to (10-n) ml of M zirconium - Calcein blueApril, 19711 283 blue solution (curves B and B', respectively) and an exactly formulated 1 : 1 zirconium - Calcein blue solution with a 200-fold molar excess of sulphate ion (curves C and C', respec- tively), all at pH 1-9.These spectra are not corrected for the spectral response charac- teristics of the gratings of the monochromators or of the photomultiplier, nor do they account for changes in the lamp emission with wavelength. As in the previously reported reaction with fluoride, sulphate ions enhance the fluorescence of the zirconium-Calcein blue complex, but produce no changes in the wavelengths of excitation or fluorescence maxima. Fig. 2 shows the effects of varying the pH of the reaction of a solution containing Calcein blue, zirconium and potassium sulphate (curve A).Curve B is for an exactly similar solution but without sulphate ions. The fluorescence of the sulphate-enh anced system reaches a maximum at pH 1-9. The pH adjustments were made by the addition of small amounts of concentrated ammonia solution. The fluorescence intensities of solutions with pH values greater than 3 are not given because their fluorescence decreases rapidly on standing. It may, however, be relevant to observe that the fluorescence intensities of the zirconium - Calcein blue solution and the sulphate-enhanced solution increases with increasing pH above 3 when measurements are made immediately. The effect of varying the concentration of Calcein blue is shown in Fig. 3. Curve A is for zirconium and potassium sulphate, curve B is for a similar solution without sulphate and curve C represents the difference between the two solutions, i.e., the net effect of the sulphate ion.It is apparent that the optimum ratio of Calcein blue to zirconium is about 1.3. This is the same ratio as that previously found for the reaction with fluoride6 and can be interpreted as evidence for a true ratio of 1 : 1 because of the low assay of the Calcein blue, which could not be obtained or purified to any greater extent in these studies. A continuous variation graph was made (Fig. 4) to investigate the probable nature of a ternary complex formed between sulphate, zirconium and Calcein blue. This was done by varying a 10-2 M sulphate solution against an exactly formulated M solution of 1 : 1 zir- conium-Calcein blue in the usual way.Such a graph requires the use of approximately equimolar solutions of zirconium - Calcein blue and sulphate but, as explained previously, the analytical procedure requires the presence of about a 100-fold molar excess of sulphate ion to obtain reasonable sensitivity. Consequently it was not possible to obtain a sharp maximum. The broad maximum obtained suggests the probable existence of a 1 : 1 : 1 ternary complex, but cannot be interpreted with any confidence as definite evidence. THE DETERMINATION OF SMALL AMOUNTS OF SULPHATE ION .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . .. .. .. . . TABLE I INTERFERENCE EFFECTS Molar excess in brackets Percentage change Ion added . . .. .... . . .. .. .. . . .. .. .. . . .. .. . . . . .. .. .. .. .. .. . . 0 0 - 58 - 46 0 Quenched - 19 0 Quenched - 75 - 56 - 16 0 0 - 77 - 63 - 46 - 45 - 37 -8 + 67 0 + 33 0 Sulphite Fluoride Aluminium Beryllium Arsenic(V) Nickel Copper (I I) Lead Calcium Cobalt Zinc Cadmium Magnesium Iron (I1 I) (100) . . (10) . . (1) - * (0.1) . . (0.01) . . (100) . . (100) .. (100) 1 . (10) .. (1) ' - (100) . * (100) . . (100) .. (100) . . (100) . . (10) . . (1) * * (100) . . (100) . . (100) . . (100) . . (1) ' . Percentage change .. .. .. .. .. .. . . .. .. . . . . .. .. . . . . . . . . . . .. .. .. .. + 100 + 52 + 13 + 100 + 14 0 0 - 42 -7 0 0 0 0 0 - 76 - 49 0 0 0 0 - 72 - 50284 TAN AND WEST: A SPECTROFLUORIMETRIC METHOD FOR [Analyst, Vol. 96 The effects of foreign ions on the determination of sulphate ion by this procedure are shown in Table I.Each solution contained 4 mg of sulphate ion in a final volume of 100 ml. The solutions were prepared by the recommended procedure and the foreign ions were added in 100-fold, 10-fold or equivalent amounts to the sulphate solution before the addition of the other reagents. It will be seen that the only serious cationic interference observed in the range of metals studied is that caused by iron(II1). None of the comnion anions interferes, except those which form more stable complexes than sulphate with the zirconium ion, e.g., oxalate and tartrate, or which tend to form insoluble compounds, e.g., tungstate and phos- phate. These anions cause low results whereas fluoride, exceptionally, gave rise to a positive result.Large amounts of fluoride, however, break down the zirconium - Calcein blue complex by formation of ZrFG2-, etc. The positive interference caused by sulphite, and to a lesser extent thiosulphate, can almost certainly be attributed to oxidation to sulphate in these dilute solutions and are not regarded as reaction of these ions per se. EXPERIMENTAL REAGENTS- Potassium sulphate solution, 10-2 M. Zirconium oxychloride solution, M-Prepare by dissolving 0.032 2 g of ZrOC1,.8H20 in 100 ml of 3 M hydrochloric acid. Prepare a M solution with 3 M hydrochloric acid. M-Dissolve 0.016 g of Calcein blue in a few drops of 0.1 M potassium hydroxide and dilute to 500ml with distilled water. The solution must be dis- carded after 2 toe3 days. Ammonia solution, 8 per cent.APPARATUS- Fluorescence measurements were made with a double monochromating spectrofluorimeter (Farrand Optical Co. Catalogue No. 104244) fitted with a 150-W Xenon arc lamp (Hanovia Division Catalogue No. 901 C-1) and an RCA IP28 photomultiplier. A Honeywell Brown recorder was used in conjunction with the spectrofluorimeter. Fused quartz cells (10 x 20 x 50 mm) were used throughout and a pH meter was used to adjust the pH. PROCEDURE- Calibration graph (2 to 12 mg or 20 to 12 000 p.p.m.)-Transfer 2 to 12 ml of 10-2 M potassium sulphate solution at suitable volume intervals into a series of 100-ml calibrated flasks and add, in the following order, 2 ml of loe4 M Calcein blue, 3 to 4 ml of 8 per cent. ammonia solution and 2 ml of Make up to the mark.Measure the fluorescence of the solutions at 410nm with an excitation wavelength of 350 nm. Deduct the fluorescence of the blank solution containing all of the reagents except the sulphate solution. These measurements can be made immediately or within 4 hours of preparation. As the volume of reagents added is less than 10m1, unknown test solutions containing down to 20 p.p.m. of sulphate can be measured if 90 ml of sample are available, or a solution as concentrated as 12 000 p.p.m., if the test aliquot is restricted to 1 ml. Calibration graph (200 to 1 000 pg or 2 to 1 000 P.p.m.)-Repeat the above procedure by using 2 to 10 ml of M Calcein blue, 3 to 4 ml of 8 per cent. ammonia solution and 2 ml of M zirconium oxychloride in 3 M hydrochloric acid. In this instance solutions as dilute as 2 p.p.m.can be analysed if 90 ml of test solution are available. M solution by 10-fold dilution of the Calcein blue solution, M zirconium oxychloride in 3 M hydrochloric acid. M potassium sulphate solution, 2 ml of CONCLUSIONS A fluorimetric method has been established for the determination of sulphate ion in the range 200 pg to 12 mg, or concentration range 2 to 12000 p.p.m., assuming a sample solution availability of 90 and 1 ml, respectively. The method is rapid and simple and has a repro- ducibility equal to or better than 5 per cent. Phosphate, oxalate and tartrate cause low recoveries and should be absent. Fluoride in small amounts produces a much greater sensitisation,6 gives rise to high results and must be absent. Large Tungstate also interferes.April, 19711 285 amounts of fluoride destroy the fluorescence of the zirconium - Calcein blue complex com- pletely.Sulphur species, which are easily oxidised in aqueous solution, interfere by sulphite formation, but the procedure shows a high tolerance towards the cations examined, except cobalt(I1) and iron(II1). The mechanism whereby the increase in fluorescence is produced is not apparent from these experiments, and although ternary complex formation may possibly be involved, the complex is too weak to allow unequivocal evidence to be adduced. The analytical method is moderately sensitive (lower limit of 2 p.p.m.) and it has the fairly wide range characteristic of many fluorimetric procedures. It furnishes a potentially useful spectroscopic analytical procedure in addition to those very few which are currently available for the determination of sulphate ion. We are grateful to the Agricultural Research Council for the provision of a grant in support of this work, and to the Science Research Council for the provision of the spectro- fluorimet er . REFERENCES THE DETERMINATION OF SMALL AMOUNTS OF SULPHATE ION 1. 2. 3. 4. 5. 6. Bertolacini, R. J., and Barney, J. E., Analyt. Chem., 1957, 29, 281. Guyon, J. C., and Lorah, E. J., Ibid., 1966, 38, 155. Jones, A. S., and Letham, D. S., Analyst, 1956, 81, 15. Hems, R. V., Kirkbright, G. F., and West, T. S., Talanta, 1969, 16, 789. Hems, R. V., Ph.D. Thesis, Imperial College, 1969. Tan, Lay Har, and West, T. S., Analyt. Chem., in the press. Received September 7th, 1970 Accepted November 25th, 1970