Analyst, May, 1973, Vol. 98, $9. 325-328 325 Titration of Sulphate in Mineral Waters and Sea Water by Using the Solid-state Lead Electrode BY MARC0 MASCINI (Istituto di Chimica Analitica, Universith di IZoma, Rome 00185, Italy) A procedure is described for the determination of sulphate in the range 20 to 3000 p.p.m. in mineral and sea waters by using a lead-selective electrode. Chloride and hydrogen carbonate are separated from the sample by passing it firstly through a cation-exchange resin in the silver form, and secondly through a cation-exchange resin in the acid form. The solution recovered is titrated with standard lead nitrate solution. THE recent availability of a lead-selective electrode has facilitated the potentiometric deter- mination of sulphate by direct titration with a standard lead solution.This method was described by Ross and Frantl and the titration was adapted by Selig to the determination of sulphur in organic compounds after oxygen-flask combustion.2 Several anions, at a high concentration, are reported to interfere in this determination; among these the anions C1- and HCO, are most commonly present at a high level in the majority of mineral waters and sea water and limit the usefulness of the method in routine applications. Sulphate was separated from other materials, both organic and inorganic, by a cation- exchange resin supported on aluminium oxide.3 Recently, details of the separation of nitrate from chloride and hydrogen carbonate by using two cation-exchange resins, with which the sample was mixed consecutively, have been published.* The first resin, in the silver form, removes the chlorides and the second, in the acid form, removes the silver ions left in the sample by the first.At the exit from the second column the sample is acidic and hydrogen carbonate is eliminated. In the work described in this paper, the feasibility of separating sulphate from chloride and hydrogen carbonate by means of ion-exchange resins was established, making possible the use of the lead-selective electrode for the titration of sulphate in most mineral and saline waters. APPARATUS AND REAGENTS- was used. was used. 1 M sodium nitrate solution. pH meter-An instrument capable of giving a scale expansion of 1 mV per division Lead ion selective electrode-A Sens Ion, Model 201 Pb, made by AMEL, Milan, Italy, Reference electrode-This had a double junction, the outer chamber being filled with Titration cell-A polyethylene beaker was used for this purpose. Magnetic stirrer.Ion-exchange resin-Merck ion exchanger I (a strongly acidic cation exchanger) was used, Analytical-reagent grade reagents were used throughout. PROCEDURES PREPARATION OF THE CATION-EXCHANGE RESIN IN THE SILVER FORM- Wash about 50 g of the analytical-grade resin with about 200 ml of 2 M nitric acid and then remove the acid by filtration on a sintered-glass funnel, washing the solid with distilled water until neutral. Next, transfer the resin to a 500-ml flask, add 200 ml of 0.25 M silver nitrate solution, stirring magnetically for about 30 minutes, and then filter the resin from the liquid on a sintered-glass funnel and wash it several times with distilled water to displace the excess of Ag+ ions.Store the resin in a dark-coloured vessel away from direct light. A 50-g portion of resin is sufficient for at least fifty analyses. @ SAC and the author.326 [Autalyst, Vol. 98 PREPARATION OF THE CATION-EXCHANGE RESIN IN THE ACID FORM- Wash 50 g of the analytical-reagent grade resin with 200 ml of 2 M nitric acid and then remove the acid by filtration on a sintered-glass funnel, washing the solid with distilled water until neutral. Use the resins in bead or column form; in the latter case prepare a column about 15 cm in height and 0.5 cm in diameter. ANALYTICAL PROCEDURE- Pass 50 to 100 ml of the sample, containing 1 to 100 mg of chloride, at 1 to 3 ml min-1 through two columns in series, the first containing the resin in the silver form, and the second in the acid form.Discard the first 10 ml. The solution is now free from chloride, hydrogen carbonate and silver ions and is slightly acidic, and should be adjusted to a pH of 5 to 6 by adding a few drops of 1 0 - 2 ~ sodium hydroxide solution and internal indicator. The pH should now be at the optimum for the lead-selective electrode.6 Alternatively, place 50 to 100 ml of the sample in a beaker with 0.5 to 1.0 g of resin in the silver form and stir magnetically for about 20 to 30 minutes. Filter the mixture through a sintered-glass funnel and, to the clear liquid, add 0.5 to 1-Og of resin in the acid form, stirring magnetically for about 20 to 30 minutes.Then, filter the solution from the resin and adjust the pH to 5 to 6. Dilute an aliquot (e.g., 10 ml, measured with a pipette) with an equal volume of 1,4-dioxan. Insert the lead-selective and reference electrodes into the magnetically stirred solution and titrate it with lead nitrate solution ; the concentration of the lead nitrate solution should be about ten times that of the sulphate. Record and plot the potential values on a graph versus the volume of titrant added to reach the end-point. The procedure does not require that the sample should be completely recovered, but that the volume taken for the final titration should be accurately known. With an unknown sample a rough titration could be carried out first on a 10-ml aliquot in order to determine the concentration of titrant required and the approximate end-point.A second 10-ml aliquot would then be used for a more accurate titration in which the titrant was added in smaller increments in the vicinity of the end-point in order to increase the precision. RESULTS AND DISCUSSION The determination of sulphate in mineral and waste waters is frequently necessary. Fig. 1 shows a typical graph of the titration of a sample containing chloride and sulphate, MASCINI: TITRATION OF SULPHATE IN MINERAL WATERS AND > E --. W Pb (NO,),/ml Fig. 1. Titration of 50 p.p.m. of sulphate ion in the presence of 100 p.p.m. of chloride ion in 1,4- dioxan (50 per cent.) with 5 x lo-3rd lead nitrate solution. Sample volume 10 ml: A, without treat- ment ; and B, with the ion-exchange treatment described in the textMay, 19731 SEA WATER BY USING THE SOLID-STATE LEAD ELECTRODE 327 at levels normally found in mineral waters, with lead nitrate solution before and after passing the sample solution through the ion-exchange columns.The advantage of using the ion- exchange process is evident. TABLE I TITRATION OF STANDARD SOLUTIONS OF SULPHATE IN THE PRESENCE OF CHLORIDE IONS Sulphate taken, p.p.m. Chloride taken, p.p.m. Sulphate found, p.p.m. 100 35 100 100 98 98 100 1000 350 980 970 1000 1020 1000 100 3500 98 100 98 98 98 1040 1040 1000 1020 1020 1000 3500 Error, p.p.m. Nil Nil 2 2 Nil 20 30 Nil 20 Nil 2 Nil 2 2 2 40 40 Nil 20 20 Phosphates are occasionally present in mineral waters and interfere in the determination of sulphate.By passing the sample through the resin in the silver form the amount of phosphate is reduced, but not eliminated; in this instance, more complicated procedures for separation must be considered.6 Tables I and I1 show the results of several titrations carried out on standard solutions of sulphate and on sea water and mineral waters. The extent of the error indicates that the TABLE I1 SULPHATE TITRATIONS IN DIFFERENT WATER SAMPLES Sulphate added Size of (as sample/ Na,SO,), Water sample. ml p.p.m. Laboratory water 10 Nil Nil Nil Nil 10 10 20 20 30 50 100 10 Nil Nil Nil Nil 10 10 20 20 30 60 100 Mineral water sample 1 Sulphate recovered (as Na,SO,), p .p . m. 35 35 40 40 45 45 55 55 60 85 140 75 70 75 77 85 85 95 90 105 125 170 Water sample Mineral water sample 2 Sea water sample 1 Sea water sample 2 Sulphate added Size of (as sample/ Na,SO,), ml p.p.m.10 Nil Nil Nil Nil 10 10 20 20 30 50 100 1 Nil Nil Nil Nil 1 Nil Nil Nil Nil Sulphate recovered (as Na,SO,), p. p. In. 50 55 50 50 60 60 75 70 85 100 150 2350 2300 2300 2300 2500 2400 2500 2450328 MASCINI procedure is acceptable for routine analysis. The limit of sensitivity is about 10 p.p.m., and the amount of chloride can be up to ten times the amount of sulphate. The presence of nitrate in the sample does not affect the results when using this procedure, at least up to the level normally found in the mineral waters, and lead nitrate solution is recommended for the titration, as it is more readily available than lead(I1) perchlorate, recommended by Ross and Frant.l The time taken to carry out the titration is about 10 minutes.In the analysis of samples of low concentration (20 to 50 p.p.m.) the electrode response is sluggish near the equivalence point, as in other potentiometric titrations. In this instance it is advisable to add the titrant in small increments and to record the value after a delay (2 minutes). In this way the titration will be more reproducible. Another valuable technique is the automatic recording of the titration volume. The resins can be regenerated several times; the first, in the silver form, can be treated with 3 M ammonia solution and washed with distilled water and then with nitric acid, as described under Preparation of the cation-exchange resin in the silver form. If the resin turns brown in colour, the ion-exchange activity becomes very low and it is necessary to change the resin. This effect is probably caused by oxidation of the resin by Ag+ ions. The resin in the acid form can be regenerated by treating it with 1 M nitric acid. REFERENCES 1. 2. 3. 4. 5. 6. Ross, J. W., jun., and Frant, M. S., Analyt. Chem., 1969, 41, 967. Selig, W., Mikrochim. Acta, 1970, 168. Kirsten, W. J., Hansson, K., and Nilsson, S. K., Analytica Chim. A d a , 1963, 28, 101. Paul, J. L., and Carlson, R. M., J . Agric. Fd Chem., 1968, 16, 766. Mascini, M., Analytica Chim. Acta, in the press. Colson, A. F., Analyst, 1963, 88, 26. Received January 27th, 1972 Amended June 30th, 1972 Accepted January 8th, 1973