412 Analyst, June, 1973, Vol. 98, $9. 412415 Determination of Chloride in Aqueous Soil Extracts and Water Samples by Means of a Chloride-selective Electrode BY A. R. SELMER-OLSEN (Chemical Research Laboratory, Agricultural University of Norway, 1432 AS-NLH, Norway) AND A. 0IEN (Institute of Soil Science, Agricultural University of Norway, 1432 As-NLH, Norway) The chloride contents of water samples and soil extracts have been determined with an Orion chloride-selective electrode. The chloride content, at levels up to 100 mg l-l, is determined in a solution that is 0.5 M with respect to ammonium nitrate and 0.03 M with respect to nitric acid. Comparison of this method with a colorimetric AutoAnalyzer method showed no significant difference between the results obtained, but for water samples low in chloride (less than 3 mgl-l) the colorimetric method was more accurate.CURRENTLY, the most widely used methods for the determination of chlorides are the gravimetric method (precipitation as silver chloride) and the argentimetric and mercurimetric methods (in which different types of coloured or potentiometric indicators are used). Bremnerl recommends the well known Mohr titrimetric method for the determination of chloride in aqueous soil extracts. This method, however, is not very sensitive, and is also subject to errors caused by adsorption effects and over-titration. Davey and Bembrickz used a silver - silver chloride electrode for the determination of chloride in aqueous extracts of soil. A potentiometric determination is much more sensitive than Mohr’s method, and is therefore more suitable for determining small concentrations of chloride in soil extracts and water samples.Technicon3 recommend a colorimetric method based on the release of thiocyanate ion from mercury(I1) thiocyanate by an equivalent amount of chloride. A red colour is formed by the reaction of thiocyanate with iron(II1). This method is very sensitive and is well suited for large numbers of samples. The Orion chloride-selective electrode appears to be convenient for the determination of chloride in soil extracts and water samples as it is more sensitive than Mohr’s method and is portable, thus enabling measurements to be made in the field. In this investigation, different types of soil and water samples have been analysed by this electrode method, and the results are compared with those obtained by the Technicon AutoAnalyzer method.EXPERIMENTAL APPARATUS- An ion-specific meter, Orion Research, Model 401, with a chloride-selective electrode, Model 94-17, was used for the potentiometric determination of chloride in water and soil extracts. The measurements were made against the Orion double-junction reference electrode, Model 90-02. A Technicon AutoAnalyzer, with a manifold system according to O’Brien,* was used for the colorimetric determination of chloride in water and soil extracts. Its use is becoming more widespread. REAGENTS- All reagents used were of analytical-reagent grade. Standard solutions of chloride-A stock solution of chloride was prepared by dissolving 2.103 g of potassium chloride in water and diluting the solution to 1 litre so as to give a 0 SAC and the authors.SELMER-OLSEN AND 0 I E N 413 concentration of 1 g 1-1 of chloride.Standard solutions in the range 0 to 100 mg 1-1 of chloride were made by diluting appropriate volumes of the stock solution with ammonium nitrate and nitric acid solutions. The final concentrations of ammonium nitrate and nitric acid were 0.5 and 0.03 M, respectively. Ammonium nitrate solution, 0.5 M-Ammonium nitrate (40 g) was dissolved in distilled water and the solution was made up to 1 litre. Nitric acid, 0.3 M-concentrated nitric acid was diluted appropriately with distilled water. Ammonium nitrate - nitric acid reagent-A solution was prepared that was 5 M in am- monium nitrate and 0.3 M in nitric acid.PROCEDURE- SoiZ samples-A 10-g amount of air-dried soil (dried a t 30 "C and passed through a 2-mm sieve) was shaken for 15 minutes with 50 ml of 0-5 M ammonium nitrate solution. The extracts were either centrifuged, or filtered through Schleicher and Schull No. 589 white ribbon filters; 2 ml of 0-3 M nitric acid were added to 20 ml of the extract before measurement with the electrode. The measured values were then compared with those given by the standard graph. Water samples-Ammonium nitrate - nitric acid reagent (2 ml) was added to 20 ml of water and the chloride content was measured with the electrode. EXTRACTION TIME- When four soils of various textures were shaken with water or 0.5 M ammonium nitrate solution for periods of 5, 30 and 60 minutes, the chloride values obtained after shaking the mixtures for 5 minutes were identical with those obtained after the longer periods of shaking.Results were the same with both extractants, provided that 2 ml of 0-3 M nitric acid per 20 ml of extract were added to the 0.5 M ammonium nitrate extracts and 2 ml of ammonium nitrate - nitric acid reagent to the aqueous extracts. These results seem reasonable as the chloride ions are only lightly bound to the negative soil particles. EFFECT OF AMMONIUM NITRATE AND NITRIC ACID- According to Orion5 the presence of ammonia can cause some interference. To overcome this effect the soil was extracted with 0.5 M ammonium nitrate solution. At this high con- centration of ammonium ions, the relatively small amounts of ammonia present in the soil will cause no significant errors.Hydroxyl ions may also interfere, and concentrations greater than eighty times the chloride concentration cause errors5 Varying the pH of the 0.5 M ammonium nitrate solution (containing 10 mg 1-1 of chloride) by addition of nitric acid showed that the same results for chloride were obtained in solutions with pH values ranging from 1.6 to 5.0. The addition of 2 ml of 0.3 M nitric acid, which should allow for any carbonate extracted from calcareous soils, is advised. Davey and Bembrick2 also found that a mixture of ammonium nitrate and nitric acid increased the accuracy of the method. We used a higher level of acidity, however. RECOVERY TESTS- The recovery of known amounts of chloride, added before extraction to four soils of various textures, ranged from 99-5 to 102 per cent.The chloride content found, expressed as milligrams of chloride per 100 g of soil, was constant, at least in the range 5 to 40 g of soil per 50 ml of extracting solution. COMPARISON OF METHODS- A series of soil extracts was prepared, and the chloride contents were determined by both the electrode and the colorimetric AutoAnalyzer methods. If the extracts were highly coloured, blank values were obtained by running the samples with water instead of mercury(I1) thiocyanate reagent in the AutoAnalyzer method. Table I shows the results for thirteen soil samples treated as follows. A 10-g sample of air-dried soil was shaken with 50 ml of 0-5 M ammonium nitrate solution for 30 minutes and the mixture then centrifuged.The chloride content in the soil extract was determined by both the electrode and the Auto- Analyzer methods after nitric acid had been added to the extracts. A t-test gave no indication RESULTS AND DISCUSSION The ionic strength will also be nearly constant.414 SELMER-OLSEN AND 01EN : DETERMINATION OF CHLORIDE I N SOIL [A?ZU&St, VOl. 98 TABLE I DETERMINATION OF CHLORIDE IN SOIL EXTRACTS Chloride content in soil extract/ mg 1-1,* found by- Soil sample ignition, electrode Auto Analyzer No. Soil type per cent. PHmo method method 2 Sand 5.9 7.5 0.7 1.8 3 Sand 7.7 7-3 0.6 0.6 4 Sandy peat soil 75.3 4.6 11.0 13.0 5 Sand 5-7 5.5 1.9 0-6 6 Sand 5.9 7.5 1.4 1.2 7 Clayey sand 10.4 7.3 0.9 0.9 8 Sandy peat soil 38.0 4.6 10.0 9.4 10 Sandy clay 7.7 6-0 1.4 1.0 11 Sandy clay 8.0 5.8 1.0 0.9 Loss on - 1 Sandy peat soil 48-8 5.9 8.5 9.4 9 Sandy peat soil 32-7 4.9 5.1 5.2 12 Sandy clay 7.8 5.9 1.6 1.4 13 Sandy peat soil 26.6 7.0 7.9 8.5 The corresponding chloride contents in soil are five times greater and range from 3.0 to 65 p.p.m.of difference (P > 0.5) between the results for the electrode and the colorimetric methods. Duplicate analyses of twenty-one water samples in the range 0.5 to 18.8p.p.m., to 30ml of each of which had been added 3 ml of a solution that was 5 M and 0.3 M with respect to ammonium nitrate and nitric acid, respectively, showed no indication of difference between the results for the two methods (Table 11). TABLE I1 DETERMINATION OF CHLORIDE IN WATER SAMPLES Sample No. 1 2 3 4 6 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 PH 6.0 6.3 6.5 6-4 6.2 6.9 6-2 6.4 6.3 6.0 5.3 4.0 6.2 6.5 6.0 5.5 7-0 5.1 7.0 4.6 5.8 Specific conductance/ $3 cm-l 40 42 83 95 75 63 191 130 35 30 84 87 39 28 30 19 25 23 23 30 54 Chloride content/mg l-l, found by- I A \ electrode method AutoAnalyzer method Duplicate results Mean Duplicate results Mean & 1.9,34 2-5 2-4,2-3 2-4 1.2,2*6 1.9 1.8.1-9 1.9 8.1,8.5 8.3 7*6,7*6 7.6 10*3,10*3 10.3 9-6,9-5 9.6 8*5,8*9 8.7 6*9,6-9 6.9 5*4,5.4 5.4 5.9,5.8 5.9 17-0,18.8 17-9 16*8,16*8 16.8 11~1,12.0 11.6 10-7,10.7 10.7 2*6,2*3 2.5 1*9,1-8 1-9 0*5,1.9 1.2 1.6,1-5 1.6 0*7,1.7 1.2 1.3,1.3 1.3 16-0,16.2 16.1 17-5,17*5 17.5 1-2,2-7 2.0 2-3,2*3 2.3 2*4,2+4 2.4 2.4,2*4 2.4 2.8,2*8 2.8 2*5,2*5 2.5 2*0,1*7 1.9 1*4,1.4 1.4 2*6,2-2 2.4 1.9,2-0 2.0 1*4,1-1 1.3 1-6,1*6 1.6 1*2,1-9 1.6 1.6,l.g 1.8 1*6,2-2 1.9 2*7,2*8 2.8 4.8.4-9 4.9 4-7,4.7 4.7 SENSITIVITY AND PRECISION OF THE METHOD- As the slope of the graph of millivolts versus concentration is not constant below a concentration of 10 p.p.m.(Fig. l), the chloride content cannot be read directly from the meter. The chloride concentration can then be measured down to 0.5 p.p.m. At this level, however, the reproducibility is poor. In duplicate analyses of thirteen soil extracts in the range A calibration graph must therefore be drawn for the lower concentrations.June, 19731 EXTRACTS AND WATER BY MEANS OF A CHLORIDE-SELECTIVE ELECTRODE 415 0.6 to 11 mg 1-1 of chloride, the coefficient of variation was 4.5 per cent. For thirteen water samples with a relatively low content of chloride (0.5 to 3 mg F), the coefficient of variation was as high as 31 per cent. For higher concentrations (443 to 18.8 mg 1-1) it was found to be about 5 per cent. I I I 1 I 1 1 0 5 10 15 2c Chloride concentration, p.p.m. Fig. I. Standard graph The results indicate that for low chloride concentrations (below 3 mg 1-1 of extract) it is better to use a colorimetric AutoAnalyzer method than the electrode method if accurate measurements are needed. Above this concentration, the chloride electrode seems to be satisfactory. The chloride content in the range 10 to 100mg1-1 can be determined by direct read-out on the instrument. REFERENCES 1. 2. 3. 4. 5 . Bremner, J. M., in Black, C. A., Editor, “Methods of Soil Analysis, Part 2, Agronomy, 9,” American Davey, B. G., and Bembrick, M. J., Proc. Soil Sci. Soc. Amer., 1969, 33, 386. Technicon AutoAnalyzer Methodology, Chloride IIa Method, Technicon Controls Inc., Chauncey, O’Brien, J. E., Wastes Engng, 1962, 670. Orion Instruction Manual for Halide Electrodes, Orion Research Incorporated, Cambridge, Mass., Received October 30th, 1972 Accepted January 9tk 1973 Society of Agronomy, Madison, Wisc., 1965, p. 947. N.Y., 1960. 1967,