ANALYST, MAY 1986, VOL. 111 539 Quantitative Determination of Thiourea in Aqueous Solution in the Presence of Sulphur Dioxide by Raman Spectroscopy Heather J. Bowley, Elizabeth A. Crathorne and Donald L. Gerrard BP Research Centre, Chertsey Road, Sunbury-on-Thames, Middlesex Tw16 7LN, UK A method has been developed for the quantitative determination of thiourea in aqueous solutions acidified with sulphuric acid in the presence of sulphur dioxide. The method uses laser Raman spectroscopy and is valid in the presence of high concentrations of sulphur dioxide and inorganic salts. This is particularly important with respect to the use of such solutions for the leaching of gold from ores and concentrates. The method is simple, rapid and accurate to within + I .2% and has been applied to the study of solutions used in leaching experiments.Keywords: Thiourea determination; sulphur dioxide; Raman spectroscopy; gold ores The use of an aqueous solution of thiourea as a leaching agent for gold has been widely reported in the literature.1-3 It offers several advantages over sodium cyanide in that it operates in acidic media, forming a cationic gold complex, and has faster dissolution kinetics. However, a major disadvantage is its tendency to oxidise , eventually forming elemental sulphur and several other decomposition products. This leads to an excessive consumption of the thiourea, which generally makes its use uneconomical in the treatment of gold-bearing ores and concentrates. The addition of sulphur dioxide during the leaching process has been found to decrease the thiourea consumption consid- erably by, it is thought, preventing the oxidation process.&6 In order to evaluate the economic feasibility of a gold leaching operation based on thiourea, it is necessary to have a reliable method for determining its concentration in acidified aqueous solutions containing sulphur dioxide and a range of metal salts.The standard methods for quantitatively determining thiourea in water are: (1) titration with N-bromosuccinimide7; (2) titration with mercury(II1) nitrate8; and (3) potentiometric titration with potassium iodate.9 It was found that iron(I1) ions, present in the ores, interfere with method 1 and sulphur dioxide interferes with methods 2 and 3. An alternative method is therefore required for this type of study, and the technique of Raman spectroscopy offers some distinct advantages for this purpose.Experimental Apparatus Potentiometric titration was carried out using a Metrohm titroprocessor. The titration curve and titre volume were plotted automatically using a Knauer dual-pen recorder. All Raman spectra were recorded using an Anaspec Model 36 laser Raman spectrometer fitted with a Tracor Northern Reticon Type S intensified diode array detector. Procedures Potentiometric titration of thiourea Thiourea is oxidised by potassium iodate in acidic solution to give formamidine disulphide: 6SC(NH2)2 + KI03 + 6H+ .+ 3[SC(NH2)2]22+ + 3H20 + KI The end-point of the titration is shown by a sharp increase in potential at a platinum redox electrode. A known volume of the solution to be analysed (5-20 ml) was pipetted into 50 ml of 1 M phosphoric acid and titrated with 0.017 M potassium iodate solution using the titroprocessor.The thiourea concentration was calculated from a knowledge of the volumes of thiourea solution taken and potassium iodate solution used and the molarity of the potassium iodate solution. Using this technique a thiourea solution, containing a weighed concentration of 7.602 g 1-1 in 0.1 M sulphuric acid, was analysed nine times and gave a mean of 7.57 g 1-1 with a standard deviation of 0.05 g 1-1. Laser Raman spectroscopy The Raman spectrum of an aqueous solution of thiourea, showing the C=S stretching mode of thiourea at 735 em-1, is shown in Fig. 1; the y (OCO) mode of acetic acid at 880 cm-1 is shown in Fig.2. As Raman spectroscopy is not intrinsically quantitative, an internal standard must be used for analyses of this type. Because of the particular application to gold leaching, in this study the aqueous solution of thiourea would contain sulphuric acid, sulphur dioxide, various inorganic salts and oxidation products of thiourea. Hence, the best internal standard for this application is one with a strong Raman band that is well separated from any bands from other components likely to be present and stable in sulphuric acid. The compound chosen was acetic acid, which has a strong Raman band at 880 cm-1 (Fig. 2) that is well separated from thiourea, Reagents All of the reagents used were of analytical-reagent grade or equivalent. Solutions of thiourea were made up in sulphuric acid (0.1 M) in various concentrations between 3 and 20 g 1-1, the mass of the thiourea being determined to three decimal places.Solutions for analysis by Raman spectroscopy also contained an accurately weighed amount of acetic acid as an internal standard (see below). An approximate 1 : 1 molar ratio of acetic acid to thiourea was used. Sodium metabisul- phite was used to dope the thiourea solutions with sulphur dioxide, assuming that 1 mol of sodium metabisulphite forms 2 mol of sulphur dioxide in acidic solution. I 1 t Q, C Q Lo LT 300 400 500 600 700 800 900 1000 1100 1200 Wavenum ber shift/cm - 1 Fig. 1. Raman spectrum of aqueous thiourea540 ANALYST, MAY 1986, VOL. 111 sulphuric acid, sulphur dioxide or common anion bands..The intensity of this band was found to be similar to that of the thiourea band at 735 cm-1 (Fig. 1) and so acetic acid was added to the solutions to be analysed at approximately the same molar concentration as the thiourea. The concentration of thiourea is then obtained from the ratio of the integrated areas of the 735 and 880 cm-l bands, and the known mass of acetic acid added to the solution. The Raman technique was calibrated using a range of standard solutions (Table 1) in order to establish the relationship between the 735 and 880 cm-1 bands. It was found necessary, because of differences in response of the various component diodes of the diode array with changes in laser power, to re-calibrate with a known standard before each analysis, as the ratio of the 735 to 880 cm-1 bands could vary by as much as 5%.It was also necessary to use the same area of the detector for each measurement to increase the accuracy of the method. Leach tests Samples (50 g) of pyrite concentrate (head assay = 17 p.p.m. of gold) were leached with 117 ml of 0.1 M sulphuric acid, containing 10 g 1-1 of thiourea, at 25 "C. The concentrate was stirred at 700 rev. min-1 in a 500-ml flat-bottomed glass vessel with indented baffles to improve the mixing. The potential of the solution was monitored by means of a platinum redox electrode, the oxidation potential being maintained between 380 and 420 mV for the first 3 h by the addition of iron(II1) I I I I I I I 1 500 600 700 800 900 1000 1100 1200 1300 Wavenu m ber s hifi/cm - Fig. 2. Raman spectrum of aqueous acetic acid Table 1.Determination of the thiourea concentration with no sulphur dioxide Thiourea concentratiodg I-' Thiourea concentration Potentiometric (by mass)/g 1-1 titration Raman spectroscopy 3.087 3.1 Standard 5.022 5.1 Standard 9.868 9.9 Standard 9.960 9.9 Standard 5.02 5.0 4.96 10.00 9.9 10.02 19.98 19.9 19.83 sulphate solution. The appropriate mass of sodium metabisul- phite was added in stages over these first 3 h to give the required thiourea to SO2 molar ratio. During the last hour of the test no adjustment was made to the potential of the solution, which fell to between 233 mV (SO2 to thiourea molar ratio of 2: 1 mlm) and 368 mV (no SO2). At the end of the leach the pulp was filtered and the residue washed with 2 x 100 ml of distilled water, the filtrate and washing being combined.The total volume of the solution was then measured. The solution was analysed for thiourea by both potentio- metric titration and laser Raman spectroscopy, the latter after the addition of the appropriate amount of acetic acid as the internal standard. The solution was also analysed for gold by atomic absorption spectrometry and the gold content of the dried residue was determined by fire assay. The percentage of the total gold extracted into solution could therefore be calculated. The mass of thiourea in the feed and product solutions was calculated and the amount of thiourea consumed in the leach test expressed as the mass of thiourea consumed (in kg) per mass of pyrite concentrate used in the test (in tonnes) i.e., kg tonne-1.Two tests were carried out using the conditions described above, but in a 1-1 vessel using 100 g of concentrate and 233 ml of a 50 g 1-1 solution of thiourea in 1 M sulphuric acid. Results and Discussion Determination of the Thiourea Concentration in the Absence of Sulphur Dioxide The concentration of thiourea in five solutions containing 5.02, 10.00 and 19.98 g 1-1 of thiourea (by mass) was determined both by potentiometric titration and laser Raman spectroscopy. The results are given in Table 1. Generally good agreement was found between the amount of thiourea added and the amount found using the two analytical techniques, the maximum error being 1% for the potentiometric and 1.2% for the laser Raman technique. Determination of the Thiourea Concentration in the Presence of Sulphur Dioxide The results are listed in Table 2.Typical traces from the potentiometric titration of a 0.1 M thiourea solution containing different amounts of sulphur dioxide are shown in Fig. 3. At an SO2 to thiourea molar ratio of 1 : 2 the change in potential owing to the reaction of the sulphur dioxide with potassium iodate can be clearly seen (marked as the SO2 end-point) although in practice the end-point of this reaction is difficult to determine. Comparing the trace of thiourea alone with that of a solution containing an SO2 to thiourea molar ratio of 1 : 10, no obvious difference in curve shape is apparent. However, the calculated thiourea concentrations of the two solutions are 7.6 g 1-1 (no SO2) and 8.3 g 1-1 (with SO2). Thus, at low concentrations of SO2 its reaction with potassium iodate cannot be detected from the shape of the potentiometric curve.The fact that a falsely high figure for the thiourea concentration is obtained is not apparent using this method at low sulphur dioxide concentrations. Table 2. Determination of thiourea concentration in the presence of sulphur dioxide Thiourea Thiourea concentratiodg 1-* (by mass)/ concentration/ SOz: thiourea Potentiometric Raman concentration Na2S205 g I-' g I-' molar ratio titration spectroscopy 10.10 12.6 1 : l 23.2,24.1 9.95 10.10 4.6 1 : 2.7 14.6,15.1 9.98 10.10 2.5 1 : 5 13.0,12.5 10.11ANALYST, MAY 1986, VOL. 111 541 No so2 Thiourea: S02=10: 1 mlm Thiourea: S02=2: 1 mim Fig. 3. Potentiometric titration of a 7.61 g 1-1 of thiourea solution Table 3.Leaching of a pyritic gold ore using thiourea Thiourea in feed Test No. (by mass)/g 1 - 1 1 10.12 2 10.12 3 10.00 4 10.00 5 50.03 6 50.03 Na2S205 addedg 0 0.75 1.4 2.8 0 8.0 SO2: thiourea Gold extraction, molar ratio % 73 1 : 2 61 1 : l 62 2 : 1 59 50 1 : 2 67 - - Table 4. Analysis of leach liquors Thiourea consumption*/ kg tonne - 1 Thiourea in feed Thiourea in leach Thiourea in leach SO,: thiourea by potentiometric liquor by potentiometric liquor by Potentiometric Raman Test No. molar ratio titration/g 1 - 1 titratiodg 1-1 Raman spectroscopy/g 1 - 1 titration spectroscopy 1 - 10.0 2.3 2.27 10.2 9.8 2 1 : 2 10.0 3.1 3.35 5.4 4.0 4 2 : 1 9.9 7.1 3.34 (-20.1) 2.8 5 - 49.8 22.5 22.55 25.1 24.9 3 1 : l 9.9 4.1 3.84 (-0.4) 1.0 6 1 : 2 49.8 23.0 22.05 (-0.5) 4.3 * Parentheses indicate that the thiourea concentration determined is apparently greater than the initial concentration of thiourea.A 0.1 M sulphuric acid solution containing 10.10 g 1-1 of thiourea was doped with different amounts of sodium metabi- sulphite to give calculated SO2 to thiourea molar ratios of 1 : 1, 1 : 2.7 and 1 : 5. The thiourea concentration of the solutions was then determined by laser Raman spectroscopy and potentiometric titration. Based on the total volume of potassium iodate used to titrate the samples, thiourea concentrations df 24, 15 and 13 g 1-1, respectively, were obtained by potentiometric titration. In each instance the presence of SOz in the solution was obvious from the shape of the potentiometric curve.The thiourea concentrations obtained for the same solutions by Raman spectroscopy were 9.95, 9.98 and 10.11 g 1-1, respectively, showing that the presence of SO2 did not affect this technique. Results of the Leaching Studies The results of the leaching studies are given in Tables 3 and 4. As the amount of sodium metabisulphite added during the leaching of the pyrite concentrate was increased, the differ- ence between the thiourea concentration of the leach liquors, as determined by the two analytical methods, also increased. With no sulphur dioxide present both methods gave similar thiourea concentrations in the leach liquor, i.e. , 2.27 g 1-1 by Raman spectroscopy and 2.3 g 1-1 by potassium iodate titration. The amount of thiourea consumed was, therefore, similar, 9.8 and 10.2 kg tonne-1, respectively. At an SO2 to thiourea molar ratio of 1 : 2 the thiourea concentration in the leach liquor was 3.35 g 1-1 by Raman spectroscopy and 3.1 g 1-1 by potassium iodate titration, giving thiourea consumptions of 4.0 and 5.4 kg tonne-1, respectively.Increasing the amount of SOz in the leach to a 1 : 1 SOz to thiourea molar ratio and above obviously left some free SO2 in the leach liquor as the analysis by potentiometric titration gave a higher thiourea concentration in this solution than the feed concentration. Analysis using Raman spectroscopy gave542 ANALYST, MAY 1986, VOL. 111 thiourea consumptions of 1.0 kg tonne-’ (SO2 to thiourea, 1 : 1) and 2.8 kg tonne-1 (SO2 to thiourea, 2 : 1). Leach tests using 50 g 1-1 of thiourea in 1 M sulphuric acid gave a greater decrease in the thiourea consumption when sulphur dioxide was present than the tests that used 0.1 M sulphuric acid.With no sulphur dioxide present, the Raman and potentiometric methods again gave similar thiourea consumption figures, 24.9 and 25.1 kg tonne-1, respectively. With an SO2 to thiourea molar ratio of 1 : 2 in the leach, the calculated value for thiourea consumed fell to 4.3 kg tonne-’ by Raman spectroscopy, whereas the potentiometric result indicated the production of 0.5 kg tonne-’, obviously an unacceptable result. The presence of sulphur dioxide in the leach tests using 0.1 M sulphuric acid appeared to decrease the gold extraction. At present, the reason for this is not apparent, but it may relate to the redox potential being allowed to fall over the final hour of the leach in order to minimise the amount of sulphur dioxide left in solution. The assistance of Mr. K. R. Nunn and advice from Mr. R. D. Hancock are gratefully acknowledged. Permission to publish this paper has been given by the British Petroleum Co. plc. References 1. 2. 3. 4. 5. 6. 7. 8. 9. Groenwald, T., Hydrometallurgy, 1976, 1, 277. Hiskey, J. B., Miner. Metall. Process., 1984, 11, 173. Pyper, R. A., and Hendrix, J. L., “Gold and Silver-Leaching, Recovery and Economics,” SME-AIME, 1981, p. 93. Lesoille, M., Br. Pat. 2077247, 1983. Schultze, R. G., J. Met., 1984, 6 , 62. Schultze, R. G., Ger. Pat., 3401 961, 1984. Thibert, R. J . , and Sarwar, M., Microchem. J., 1969, 14,271. Yatsimirsky, K. B., and Artasheva, A. A., Zh. Anal. Khim., 1956, 11, 442. Schultze, R. G., personal communication. Paper A51363 Received October 11 th, 1985 Accepted November 25th, 1985